├── data ├── mask_dataset.names └── mask_dataset_validate.txt ├── .gitignore ├── testing ├── screenshots │ ├── ss1.png │ ├── ss2.png │ └── THUMBNAIL.png ├── input │ └── images │ │ ├── img.jpg │ │ └── image2.jpg ├── output │ └── images │ │ ├── img.jpg │ │ └── image2.jpg └── ExplanantionImg │ ├── nms.jpg │ ├── 10647.jpg │ ├── final.jpg │ ├── cls_prediction.jpg │ └── thresholding_2.jpg ├── requirements.txt ├── config ├── mask_dataset.data └── yolov3_mask.cfg ├── explanation.md ├── utils ├── parse_config.py ├── datasets.py └── utils.py ├── README.md ├── validate.py ├── train.py ├── image_detect.py ├── cam_detect.py ├── video_detect.py └── models.py /data/mask_dataset.names: -------------------------------------------------------------------------------- 1 | MASK 2 | NO_MASK 3 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | 2 | checkpoints/*.pth 3 | .idea 4 | utils/__pycache__ 5 | __pycache__ 6 | .gitignore -------------------------------------------------------------------------------- /testing/screenshots/ss1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/screenshots/ss1.png -------------------------------------------------------------------------------- /testing/screenshots/ss2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/screenshots/ss2.png -------------------------------------------------------------------------------- /testing/input/images/img.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/input/images/img.jpg -------------------------------------------------------------------------------- /testing/output/images/img.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/output/images/img.jpg -------------------------------------------------------------------------------- /testing/ExplanantionImg/nms.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/ExplanantionImg/nms.jpg -------------------------------------------------------------------------------- /testing/input/images/image2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/input/images/image2.jpg -------------------------------------------------------------------------------- /testing/output/images/image2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/output/images/image2.jpg -------------------------------------------------------------------------------- /testing/ExplanantionImg/10647.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/ExplanantionImg/10647.jpg -------------------------------------------------------------------------------- /testing/ExplanantionImg/final.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/ExplanantionImg/final.jpg -------------------------------------------------------------------------------- /testing/screenshots/THUMBNAIL.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/screenshots/THUMBNAIL.png -------------------------------------------------------------------------------- /testing/ExplanantionImg/cls_prediction.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/ExplanantionImg/cls_prediction.jpg -------------------------------------------------------------------------------- /testing/ExplanantionImg/thresholding_2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/NisargPethani/FACE-MASK-DETECTION-USING-YOLO-V3/HEAD/testing/ExplanantionImg/thresholding_2.jpg -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | numpy==1.18.1 2 | pillow>=6.2.2 3 | torch==1.4.0 4 | tqdm 5 | terminaltables 6 | torchvision 7 | matplotlib 8 | tensorboard 9 | opencv-python 10 | -------------------------------------------------------------------------------- /config/mask_dataset.data: -------------------------------------------------------------------------------- 1 | classes=2 2 | train=data/mask_dataset_train.txt 3 | valid=data/mask_dataset_validate.txt 4 | names=data/mask_dataset.names 5 | backup=backup/ 6 | eval=coco 7 | 8 | -------------------------------------------------------------------------------- /explanation.md: -------------------------------------------------------------------------------- 1 | # Explanation 2 | 3 | ![final.jpg](testing/ExplanantionImg/final.jpg) 4 | 5 | - Object Detection Process 6 | - **Localization** 7 | - **Class Prediction** 8 | 9 | ![10647.jpg](testing/ExplanantionImg/10647.jpg) 10 | 11 | - Thresholding 12 | 13 | ![thresholding 2.jpg](testing/ExplanantionImg/thresholding_2.jpg) 14 | 15 | - Non max suppression irrespective of class label 16 | 17 | ![nms.jpg](testing/ExplanantionImg/nms.jpg) 18 | 19 | - Bounding Box Labelling 20 | 21 | ![cls prediction.jpg](testing/ExplanantionImg/cls_prediction.jpg) -------------------------------------------------------------------------------- /utils/parse_config.py: -------------------------------------------------------------------------------- 1 | 2 | 3 | def parse_model_config(path): 4 | """Parses the yolo-v3 layer configuration file and returns module definitions""" 5 | file = open(path, 'r') 6 | lines = file.read().split('\n') 7 | lines = [x for x in lines if x and not x.startswith('#')] 8 | lines = [x.rstrip().lstrip() for x in lines] # get rid of fringe whitespaces 9 | module_defs = [] 10 | for line in lines: 11 | if line.startswith('['): # This marks the start of a new block 12 | module_defs.append({}) 13 | module_defs[-1]['type'] = line[1:-1].rstrip() 14 | if module_defs[-1]['type'] == 'convolutional': 15 | module_defs[-1]['batch_normalize'] = 0 16 | else: 17 | key, value = line.split("=") 18 | value = value.strip() 19 | module_defs[-1][key.rstrip()] = value.strip() 20 | 21 | return module_defs 22 | 23 | def parse_data_config(path): 24 | """Parses the data configuration file""" 25 | options = dict() 26 | options['gpus'] = '0,1,2,3' 27 | options['num_workers'] = '10' 28 | with open(path, 'r') as fp: 29 | lines = fp.readlines() 30 | for line in lines: 31 | line = line.strip() 32 | if line == '' or line.startswith('#'): 33 | continue 34 | key, value = line.split('=') 35 | options[key.strip()] = value.strip() 36 | return options 37 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # FACE-MASK DETECTION 2 | A minimal PyTorch implementation of YOLOv3, with support for training, interface & evalution.
3 | To train on custom dataset please visit my another [[GitRepo]](). 4 | 5 | ## ABSTARCT 6 | ### APPROCH 7 | Standard YOLO V3 Approch is modified, and the table shows what change has been made. 8 | 9 | |Standard YOLO Approch | Self-Modified YOLO Approch | 10 | |-----------------------------------------------|--------------------------------------------------| 11 | | 1. Object Detection Process | 1. Object Detection Process | 12 | | ... 1.1. Localization | ... 1.1. Localization | 13 | | ... 1.2. Class Prediction | ... 1.2. Class Prediction | 14 | | 2. Thresholding | 2. Thresholding | 15 | | 3. Non max suppression with respect to Class | 3. Non max suppression irrespective of class label| 16 | ||4. Bounding Box Labelling| 17 | 18 | ### REPORT 19 | - Explanation : [explanation.md](explanation.md) 20 | - You can check the full report of this project here... [[FACE-MASK DETECTION USING YOLO V3 ARCHITECTURE.pdf]](https://drive.google.com/file/d/1QFFEEtHlMsQHcgEiQYy4hlYvCam73KE5/view?usp=sharing) 21 | 22 | 23 | ## INSTALLATION 24 | ##### Clone and install requirements 25 | > git clone https://github.com/NisargPethani/Face-Mask-Detection-using-YOLO-v3.git 26 | > cd Face-Mask-Detection-using-YOLO-v3/ 27 | > pip install -r requirements.txt 28 | 29 | 30 | ##### Checkpoints 31 | Download checkpoint From [[GoogleDrive]](https://drive.google.com/drive/folders/1UlF6PmTwwd4cm-wD9v6Qy7gbC_tzif_j?usp=sharing)
32 | Copy `yolov3_ckpt_35.pth` into `checkpoints/` 33 | 34 | 35 | ## EVALUATION 36 | Evaluates the model. 37 | 38 | > python validate.py --weights_path checkpoints/yolov3_ckpt_35.pth 39 | 40 | Average Precisions: 41 | | Class | AP | 42 | | ----------------------- |:-----------------:| 43 | | Class '0' (MASK) | 73.0 | 44 | | Class '1' (NO_MASK) | 83.3 | 45 | 46 | mAP: 78.19 47 | 48 | ## DETECTION 49 | ### Real Time Detection 50 | > python cam_detect.py --weights_path checkpoints/yolov3_ckpt_35.pth 51 | Some Screen-shots of Real-Time Detection is shown below 52 | 53 |

54 |

55 | 56 | 57 | ### Detection in Image 58 | Move inmages to `testing/input/images` 59 | 60 | > python image_detect.py --image_folder testing/input/images --weights_path checkpoints/yolov3_ckpt_35.pth 61 | 62 |

63 |

64 | 65 | 66 | ### Detection in Video 67 | Make new directory with name: `'videos'` in `testing/input`
68 | Move videos to `testing/input/videos` 69 | 70 | > python video_detect.py --image_folder testing/input/videos --weights_path checkpoints/yolov3_ckpt_35.pth 71 | 72 | 73 | ## YOUTUBE 74 | Following YouTube video shows the output. 75 |

77 | 78 | 79 | ## CREDIT 80 | [[Paper]](https://pjreddie.com/media/files/papers/YOLOv3.pdf) [[Project Webpage]](https://pjreddie.com/darknet/yolo/) [[Authors' Implementation]](https://github.com/pjreddie/darknet) 81 | 82 | ``` 83 | @article{yolov3, 84 | title={YOLOv3: An Incremental Improvement}, 85 | author={Redmon, Joseph and Farhadi, Ali}, 86 | journal = {arXiv}, 87 | year={2018} 88 | } 89 | ``` 90 | 91 | Also Help is taken from [[GitRepo]](https://github.com/eriklindernoren/PyTorch-YOLOv3.git) -------------------------------------------------------------------------------- /validate.py: -------------------------------------------------------------------------------- 1 | from __future__ import division 2 | 3 | from models import Darknet 4 | from utils.utils import non_max_suppression, non_max_suppression_output, xywh2xyxy, get_batch_statistics, ap_per_class, load_classes 5 | from utils.datasets import ListDataset 6 | from utils.parse_config import parse_data_config 7 | 8 | import numpy as np 9 | import argparse 10 | import tqdm 11 | 12 | import torch 13 | from torch.utils.data import DataLoader 14 | from torch.autograd import Variable 15 | 16 | 17 | def evaluate(model, path, iou_thres, conf_thres, nms_thres, img_size, batch_size): 18 | model.eval() 19 | 20 | # Get dataloader 21 | dataset = ListDataset(path, img_size=img_size, augment=False, multiscale=False) 22 | dataloader = torch.utils.data.DataLoader( 23 | dataset, batch_size=batch_size, shuffle=False, num_workers=1, collate_fn=dataset.collate_fn 24 | ) 25 | 26 | Tensor = torch.cuda.FloatTensor if torch.cuda.is_available() else torch.FloatTensor 27 | 28 | labels = [] 29 | sample_metrics = [] # List of tuples (TP, confs, pred) 30 | for batch_i, (_, imgs, targets) in enumerate(tqdm.tqdm(dataloader, desc="Detecting objects")): 31 | 32 | # Extract labels 33 | labels += targets[:, 1].tolist() 34 | # Rescale target 35 | targets[:, 2:] = xywh2xyxy(targets[:, 2:]) 36 | targets[:, 2:] *= img_size 37 | 38 | imgs = Variable(imgs.type(Tensor), requires_grad=False) 39 | 40 | with torch.no_grad(): 41 | outputs = model(imgs) 42 | outputs = non_max_suppression(outputs, conf_thres=conf_thres, nms_thres=nms_thres) 43 | 44 | sample_metrics += get_batch_statistics(outputs, targets, iou_threshold=iou_thres) 45 | 46 | # Concatenate sample statistics 47 | true_positives, pred_scores, pred_labels = [np.concatenate(x, 0) for x in list(zip(*sample_metrics))] 48 | precision, recall, AP, f1, ap_class = ap_per_class(true_positives, pred_scores, pred_labels, labels) 49 | 50 | return precision, recall, AP, f1, ap_class 51 | 52 | 53 | if __name__ == "__main__": 54 | parser = argparse.ArgumentParser() 55 | parser.add_argument("--batch_size", type=int, default=8, help="size of each image batch") 56 | parser.add_argument("--model_def", type=str, default="config/yolov3_mask.cfg", help="path to model definition file") 57 | parser.add_argument("--data_config", type=str, default="config/mask_dataset.data", help="path to data config file") 58 | parser.add_argument("--weights_path", type=str, default="checkpoints/yolov3_ckpt_36.pth", help="path to weights file") 59 | parser.add_argument("--class_path", type=str, default="data/mask_dataset.names", help="path to class label file") 60 | parser.add_argument("--iou_thres", type=float, default=0.4, help="iou threshold required to qualify as detected") 61 | parser.add_argument("--conf_thres", type=float, default=0.9, help="object confidence threshold") 62 | parser.add_argument("--nms_thres", type=float, default=0.4, help="iou thresshold for non-maximum suppression") 63 | parser.add_argument("--img_size", type=int, default=416, help="size of each image dimension") 64 | opt = parser.parse_args() 65 | 66 | print("") 67 | for arg in vars(opt): 68 | print(str(arg) +":\t\t\t"+ str(getattr(opt, arg))) 69 | print("") 70 | 71 | device = torch.device("cuda" if torch.cuda.is_available() else "cpu") 72 | 73 | data_config = parse_data_config(opt.data_config) 74 | valid_path = data_config["valid"] 75 | class_names = load_classes(data_config["names"]) 76 | 77 | # Initiate model 78 | model = Darknet(opt.model_def).to(device) 79 | if opt.weights_path.endswith(".weights"): 80 | # Load darknet weights 81 | model.load_darknet_weights(opt.weights_path) 82 | else: 83 | # Load checkpoint weights 84 | model.load_state_dict(torch.load(opt.weights_path)) 85 | 86 | print("Compute mAP...") 87 | print("") 88 | 89 | precision, recall, AP, f1, ap_class = evaluate( 90 | model, 91 | path=valid_path, 92 | iou_thres=opt.iou_thres, 93 | conf_thres=opt.conf_thres, 94 | nms_thres=opt.nms_thres, 95 | img_size=opt.img_size, 96 | batch_size=8, 97 | ) 98 | print("") 99 | print("Average Precisions:") 100 | for i, c in enumerate(ap_class): 101 | print(f"+ Class '{c}' ({class_names[c]}) - AP: {AP[i]}") 102 | 103 | print("") 104 | print(f"mAP: {AP.mean()}") 105 | -------------------------------------------------------------------------------- /utils/datasets.py: -------------------------------------------------------------------------------- 1 | import glob 2 | import random 3 | import os 4 | import numpy as np 5 | from PIL import Image 6 | import torch 7 | import torch.nn.functional as F 8 | 9 | from torch.utils.data import Dataset 10 | import torchvision.transforms as transforms 11 | 12 | def horisontal_flip(images, targets): 13 | images = torch.flip(images, [-1]) 14 | targets[:, 2] = 1 - targets[:, 2] 15 | return images, targets 16 | 17 | def pad_to_square(img, pad_value): 18 | c, h, w = img.shape 19 | dim_diff = np.abs(h - w) 20 | # (upper / left) padding and (lower / right) padding 21 | pad1, pad2 = dim_diff // 2, dim_diff - dim_diff // 2 22 | # Determine padding 23 | pad = (0, 0, pad1, pad2) if h <= w else (pad1, pad2, 0, 0) 24 | # Add padding 25 | img = F.pad(img, pad, "constant", value=pad_value) 26 | 27 | return img, pad 28 | 29 | 30 | def resize(image, size): 31 | image = F.interpolate(image.unsqueeze(0), size=size, mode="nearest").squeeze(0) 32 | return image 33 | 34 | class ImageFolder(Dataset): 35 | def __init__(self, folder_path, img_size=416): 36 | self.files = sorted(glob.glob("%s/*.*" % folder_path)) 37 | self.img_size = img_size 38 | 39 | def __getitem__(self, index): 40 | img_path = self.files[index % len(self.files)] 41 | # Extract image as PyTorch tensor 42 | img = transforms.ToTensor()(Image.open(img_path)) 43 | # Pad to square resolution 44 | img, _ = pad_to_square(img, 0) 45 | # Resize 46 | img = resize(img, self.img_size) 47 | 48 | return img_path, img 49 | 50 | def __len__(self): 51 | return len(self.files) 52 | 53 | 54 | class ListDataset(Dataset): 55 | def __init__(self, list_path, img_size=416, augment=True, multiscale=True, normalized_labels=True): 56 | with open(list_path, "r") as file: 57 | self.img_files = file.readlines() 58 | 59 | self.label_files = [ 60 | path.replace("images", "labels").replace(".png", ".txt").replace(".jpg", ".txt") 61 | for path in self.img_files 62 | ] 63 | self.img_size = img_size 64 | self.max_objects = 100 65 | self.augment = augment 66 | self.multiscale = multiscale 67 | self.normalized_labels = normalized_labels 68 | self.min_size = self.img_size - 3 * 32 69 | self.max_size = self.img_size + 3 * 32 70 | self.batch_count = 0 71 | 72 | def __getitem__(self, index): 73 | 74 | # --------- 75 | # Image 76 | # --------- 77 | 78 | img_path = self.img_files[index % len(self.img_files)].rstrip() 79 | 80 | # Extract image as PyTorch tensor 81 | img = transforms.ToTensor()(Image.open(img_path).convert('RGB')) 82 | 83 | # Handle images with less than three channels 84 | if len(img.shape) != 3: 85 | img = img.unsqueeze(0) 86 | img = img.expand((3, img.shape[1:])) 87 | 88 | _, h, w = img.shape 89 | h_factor, w_factor = (h, w) if self.normalized_labels else (1, 1) 90 | # Pad to square resolution 91 | img, pad = pad_to_square(img, 0) 92 | _, padded_h, padded_w = img.shape 93 | 94 | # --------- 95 | # Label 96 | # --------- 97 | 98 | label_path = self.label_files[index % len(self.img_files)].rstrip() 99 | 100 | targets = None 101 | if os.path.exists(label_path): 102 | boxes = torch.from_numpy(np.loadtxt(label_path).reshape(-1, 5)) 103 | # Extract coordinates for unpadded + unscaled image 104 | x1 = w_factor * (boxes[:, 1] - boxes[:, 3] / 2) 105 | y1 = h_factor * (boxes[:, 2] - boxes[:, 4] / 2) 106 | x2 = w_factor * (boxes[:, 1] + boxes[:, 3] / 2) 107 | y2 = h_factor * (boxes[:, 2] + boxes[:, 4] / 2) 108 | # Adjust for added padding 109 | x1 += pad[0] 110 | y1 += pad[2] 111 | x2 += pad[1] 112 | y2 += pad[3] 113 | # Returns (x, y, w, h) 114 | boxes[:, 1] = ((x1 + x2) / 2) / padded_w 115 | boxes[:, 2] = ((y1 + y2) / 2) / padded_h 116 | boxes[:, 3] *= w_factor / padded_w 117 | boxes[:, 4] *= h_factor / padded_h 118 | 119 | targets = torch.zeros((len(boxes), 6)) 120 | targets[:, 1:] = boxes 121 | 122 | # Apply augmentations 123 | if self.augment: 124 | if np.random.random() < 0.5: 125 | img, targets = horisontal_flip(img, targets) 126 | 127 | return img_path, img, targets 128 | 129 | def collate_fn(self, batch): 130 | paths, imgs, targets = list(zip(*batch)) 131 | # Remove empty placeholder targets 132 | targets = [boxes for boxes in targets if boxes is not None] 133 | # Add sample index to targets 134 | for i, boxes in enumerate(targets): 135 | boxes[:, 0] = i 136 | targets = torch.cat(targets, 0) 137 | # Selects new image size every tenth batch 138 | if self.multiscale and self.batch_count % 10 == 0: 139 | self.img_size = random.choice(range(self.min_size, self.max_size + 1, 32)) 140 | # Resize images to input shape 141 | imgs = torch.stack([resize(img, self.img_size) for img in imgs]) 142 | self.batch_count += 1 143 | return paths, imgs, targets 144 | 145 | def __len__(self): 146 | return len(self.img_files) 147 | -------------------------------------------------------------------------------- /train.py: -------------------------------------------------------------------------------- 1 | from __future__ import division 2 | 3 | from models import Darknet 4 | from utils.utils import load_classes, weights_init_normal 5 | from utils.datasets import ListDataset 6 | from utils.parse_config import parse_data_config 7 | from validate import evaluate 8 | 9 | from terminaltables import AsciiTable 10 | 11 | import os 12 | import time 13 | import argparse 14 | 15 | import torch 16 | from torch.utils.data import DataLoader 17 | from torch.autograd import Variable 18 | 19 | import warnings 20 | warnings.filterwarnings("ignore", category=UserWarning) 21 | 22 | # from utils.logger import * 23 | 24 | if __name__ == "__main__": 25 | parser = argparse.ArgumentParser() 26 | parser.add_argument("--epochs", type=int, default=100, help="number of epochs") 27 | parser.add_argument("--batch_size", type=int, default=6, help="size of each image batch") 28 | parser.add_argument("--gradient_accumulations", type=int, default=2, help="number of gradient accums before step") 29 | parser.add_argument("--model_def", type=str, default="config/yolov3_mask.cfg", help="path to model definition file") 30 | parser.add_argument("--data_config", type=str, default="config/mask_dataset.data", help="path to data config file") 31 | parser.add_argument("--pretrained_weights", type=str, default="weights/yolov3.weights", help="if specified starts from checkpoint model") 32 | parser.add_argument("--n_cpu", type=int, default=8, help="number of cpu threads to use during batch generation") 33 | parser.add_argument("--img_size", type=int, default=416, help="size of each image dimension") 34 | parser.add_argument("--checkpoint_interval", type=int, default=1, help="interval between saving model weights") 35 | parser.add_argument("--evaluation_interval", type=int, default=1, help="interval evaluations on validation set") 36 | parser.add_argument("--compute_map", default=True, help="if True computes mAP every tenth batch") 37 | parser.add_argument("--multiscale_training", default=True, help="allow for multi-scale training") 38 | opt = parser.parse_args() 39 | print(opt) 40 | 41 | # logger = Logger("logs") 42 | 43 | device = torch.device("cuda" if torch.cuda.is_available() else "cpu") 44 | 45 | os.makedirs("output", exist_ok=True) 46 | os.makedirs("checkpoints", exist_ok=True) 47 | 48 | # Get data configuration 49 | data_config = parse_data_config(opt.data_config) 50 | train_path = data_config["train"] 51 | valid_path = data_config["valid"] 52 | class_names = load_classes(data_config["names"]) 53 | 54 | # Initiate model 55 | model = Darknet(opt.model_def).to(device) 56 | model.apply(weights_init_normal) 57 | 58 | # If specified we start from checkpoint 59 | if opt.pretrained_weights: 60 | if opt.pretrained_weights.endswith(".pth"): 61 | model.load_state_dict(torch.load(opt.pretrained_weights)) 62 | else: 63 | model.load_darknet_weights(opt.pretrained_weights) 64 | 65 | # Get dataloader 66 | dataset = ListDataset(train_path, augment=True, multiscale=opt.multiscale_training) 67 | dataloader = torch.utils.data.DataLoader( 68 | dataset, 69 | batch_size=opt.batch_size, 70 | shuffle=True, 71 | num_workers=opt.n_cpu, 72 | pin_memory=True, 73 | collate_fn=dataset.collate_fn, 74 | ) 75 | 76 | optimizer = torch.optim.Adam(model.parameters()) 77 | 78 | # to get mAP 79 | to_get_mAP = None 80 | 81 | for epoch in range(opt.epochs): 82 | model.train() 83 | start_time = time.time() 84 | for batch_i, (_, imgs, targets) in enumerate(dataloader): 85 | batches_done = len(dataloader) * epoch + batch_i 86 | 87 | imgs = Variable(imgs.to(device)) 88 | targets = \ 89 | Variable(targets.to(device), requires_grad=False) 90 | 91 | loss, outputs = model(imgs, targets) 92 | loss.backward() 93 | 94 | if batches_done % opt.gradient_accumulations: 95 | optimizer.step() 96 | optimizer.zero_grad() 97 | 98 | 99 | log_str = "---- [Epoch %d/%d, Batch %d/%d] ----" % (epoch, opt.epochs, batch_i, len(dataloader)) 100 | log_str += f"Total loss {loss.item()}" 101 | print(log_str) 102 | 103 | model.seen += imgs.size(0) 104 | 105 | if epoch % opt.evaluation_interval == 0: 106 | try: 107 | print("\n---- Evaluating Model ----") 108 | # Evaluate the model on the validation set 109 | precision, recall, AP, f1, ap_class = evaluate( 110 | model, 111 | path=valid_path, 112 | iou_thres=0.5, 113 | conf_thres=0.5, 114 | nms_thres=0.5, 115 | img_size=opt.img_size, 116 | batch_size=4, 117 | ) 118 | evaluation_metrics = [ 119 | ("val_precision", precision.mean()), 120 | ("val_recall", recall.mean()), 121 | ("val_mAP", AP.mean()), 122 | ("val_f1", f1.mean()), 123 | ] 124 | # logger.list_of_scalars_summary(evaluation_metrics, epoch) 125 | 126 | # Print class APs and mAP 127 | ap_table = [["Index", "Class name", "AP"]] 128 | for i, c in enumerate(ap_class): 129 | ap_table += [[c, class_names[c], "%.5f" % AP[i]]] 130 | print(AsciiTable(ap_table).table) 131 | print(f"---- mAP {AP.mean()}") 132 | to_get_mAP = AP.mean() 133 | except: 134 | to_get_mAP = 999999999999 135 | 136 | if epoch % opt.checkpoint_interval == 0: 137 | torch.save(model.state_dict(), "checkpoints/23-04-2020__02-35/yolov3_ckpt_{0}__'{1}'__'{2}'.pth".format(epoch ,loss.item(),to_get_mAP)) 138 | -------------------------------------------------------------------------------- /image_detect.py: -------------------------------------------------------------------------------- 1 | from __future__ import division 2 | 3 | from models import Darknet 4 | from utils.utils import load_classes,non_max_suppression_output, non_max_suppression 5 | 6 | import argparse 7 | 8 | import os 9 | import torch 10 | import numpy as np 11 | from torch.autograd import Variable 12 | 13 | import cv2 14 | 15 | if __name__ == "__main__": 16 | parser = argparse.ArgumentParser() 17 | parser.add_argument("--input_file_path", type=str, default="testing/input/images", help="path to images directory") 18 | parser.add_argument("--output_path", type=str, default="testing/output/images", help="output image directory") 19 | parser.add_argument("--model_def", type=str, default="config/yolov3_mask.cfg", help="path to model definition file") 20 | parser.add_argument("--weights_path", type=str, default="checkpoints/yolov3_ckpt_35.pth", help="path to weights file") 21 | parser.add_argument("--class_path", type=str, default="data/mask_dataset.names", help="path to class label file") 22 | parser.add_argument("--conf_thres", type=float, default=0.8, help="object confidence threshold") 23 | parser.add_argument("--nms_thres", type=float, default=0.3, help="iou thresshold for non-maximum suppression") 24 | parser.add_argument("--frame_size", type=int, default=416, help="size of each image dimension") 25 | 26 | opt = parser.parse_args() 27 | print(opt) 28 | 29 | # Output directory 30 | os.makedirs(opt.output_path, exist_ok=True) 31 | 32 | # checking for GPU 33 | device = torch.device("cuda" if torch.cuda.is_available() else "cpu") 34 | 35 | # Set up model 36 | model = Darknet(opt.model_def, img_size=opt.frame_size).to(device) 37 | 38 | # loading weights 39 | if opt.weights_path.endswith(".weights"): 40 | model.load_darknet_weights(opt.weights_path) # Load weights 41 | else: 42 | model.load_state_dict(torch.load(opt.weights_path)) # Load checkpoints 43 | 44 | # Set in evaluation mode 45 | model.eval() 46 | 47 | # Extracts class labels from file 48 | classes = load_classes(opt.class_path) 49 | 50 | # ckecking for GPU for Tensor 51 | Tensor = torch.cuda.FloatTensor if torch.cuda.is_available() else torch.FloatTensor 52 | 53 | print("\nPerforming object detection:") 54 | 55 | # for text in output 56 | t_size = cv2.getTextSize(" ", cv2.FONT_HERSHEY_PLAIN, 1, 1)[0] 57 | 58 | for imagename in os.listdir(opt.input_file_path): 59 | 60 | print("\n"+imagename+"_______") 61 | image_path = os.path.join(opt.input_file_path, imagename) 62 | 63 | # frame extraction 64 | org_img = cv2.imread(image_path) 65 | 66 | # Original image width and height 67 | i_height, i_width = org_img.shape[:2] 68 | 69 | # resizing => [BGR -> RGB] => [[0...255] -> [0...1]] => [[3, 416, 416] -> [416, 416, 3]] 70 | # => [[416, 416, 3] => [416, 416, 3, 1]] => [np_array -> tensor] => [tensor -> variable] 71 | 72 | # resizing to [416 x 416] 73 | 74 | # Create a black image 75 | x = y = i_height if i_height > i_width else i_width 76 | 77 | # Black image 78 | img = np.zeros((x, y, 3), np.uint8) 79 | 80 | # Putting original image into black image 81 | start_new_i_height = int((y - i_height) / 2) 82 | start_new_i_width = int((x - i_width) / 2) 83 | 84 | img[start_new_i_height: (start_new_i_height + i_height) ,start_new_i_width: (start_new_i_width + i_width) ] = org_img 85 | 86 | #resizing to [416x 416] 87 | img = cv2.resize(img, (opt.frame_size, opt.frame_size)) 88 | 89 | # [BGR -> RGB] 90 | img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) 91 | # [[0...255] -> [0...1]] 92 | img = np.asarray(img) / 255 93 | # [[3, 416, 416] -> [416, 416, 3]] 94 | img = np.transpose(img, [2, 0, 1]) 95 | # [[416, 416, 3] => [416, 416, 3, 1]] 96 | img = np.expand_dims(img, axis=0) 97 | # [np_array -> tensor] 98 | img = torch.Tensor(img) 99 | 100 | # plt.imshow(img[0].permute(1, 2, 0)) 101 | # plt.show() 102 | 103 | # [tensor -> variable] 104 | img = Variable(img.type(Tensor)) 105 | 106 | # Get detections 107 | with torch.no_grad(): 108 | detections = model(img) 109 | 110 | detections = non_max_suppression_output(detections, opt.conf_thres, opt.nms_thres) 111 | 112 | # print(detections) 113 | 114 | # For accommodate results in original frame 115 | mul_constant = x / opt.frame_size 116 | 117 | # For each detection in detections 118 | for detection in detections: 119 | if detection is not None: 120 | 121 | print("{0} Detection found".format(len(detection))) 122 | for x1, y1, x2, y2, conf, cls_conf, cls_pred in detection: 123 | 124 | # Accommodate bounding box in original frame 125 | x1 = int(x1 * mul_constant - start_new_i_width) 126 | y1 = int(y1 * mul_constant - start_new_i_height) 127 | x2 = int(x2 * mul_constant - start_new_i_width) 128 | y2 = int(y2 * mul_constant - start_new_i_height) 129 | 130 | # Bounding box making and setting Bounding box title 131 | if (int(cls_pred) == 0): 132 | # WITH_MASK 133 | cv2.rectangle(org_img, (x1, y1), (x2, y2), (0, 255, 0), 2) 134 | else: 135 | #WITHOUT_MASK 136 | cv2.rectangle(org_img, (x1, y1), (x2, y2), (0, 0, 255), 2) 137 | 138 | cv2.putText(org_img, classes[int(cls_pred)]+": %.2f" %conf, (x1, y1 + t_size[1] + 4), cv2.FONT_HERSHEY_PLAIN, 1, 139 | [225, 255, 255], 2) 140 | 141 | 142 | 143 | out_filepath = os.path.join(opt.output_path, imagename) 144 | cv2.imwrite(out_filepath,org_img) 145 | 146 | print("Done....") 147 | 148 | cv2.destroyAllWindows() 149 | -------------------------------------------------------------------------------- /cam_detect.py: -------------------------------------------------------------------------------- 1 | from __future__ import division 2 | 3 | from models import Darknet 4 | from utils.utils import load_classes,non_max_suppression_output 5 | 6 | import argparse 7 | 8 | import time 9 | import torch 10 | import numpy as np 11 | from torch.autograd import Variable 12 | from datetime import datetime 13 | 14 | import cv2 15 | 16 | if __name__ == "__main__": 17 | parser = argparse.ArgumentParser() 18 | parser.add_argument("--model_def", type=str, default="config/yolov3_mask.cfg", help="path to model definition file") 19 | parser.add_argument("--weights_path", type=str, default="checkpoints/yolov3_ckpt_35.pth", help="path to weights file") 20 | parser.add_argument("--class_path", type=str, default="data/mask_dataset.names", help="path to class label file") 21 | parser.add_argument("--conf_thres", type=float, default=0.9, help="object confidence threshold") 22 | parser.add_argument("--nms_thres", type=float, default=0.4, help="iou thresshold for non-maximum suppression") 23 | parser.add_argument("--frame_size", type=int, default=416, help="size of each image dimension") 24 | 25 | opt = parser.parse_args() 26 | print(opt) 27 | 28 | if(torch.cuda.is_available()): 29 | print("Running on GPU") 30 | else: 31 | print("Running on CPU") 32 | 33 | # checking for GPU 34 | device = torch.device("cuda" if torch.cuda.is_available() else "cpu") 35 | 36 | # Set up model 37 | model = Darknet(opt.model_def, img_size=opt.frame_size).to(device) 38 | 39 | # loading weights 40 | if opt.weights_path.endswith(".weights"): 41 | model.load_darknet_weights(opt.weights_path) # Load weights 42 | else: 43 | model.load_state_dict(torch.load(opt.weights_path)) # Load checkpoints 44 | 45 | # Set in evaluation mode 46 | model.eval() 47 | 48 | # Extracts class labels from file 49 | classes = load_classes(opt.class_path) 50 | 51 | # ckecking for GPU for Tensor 52 | Tensor = torch.cuda.FloatTensor if torch.cuda.is_available() else torch.FloatTensor 53 | 54 | # camara capture 55 | # "http://192.168.43.19:8080/video" 56 | cap = cv2.VideoCapture(0) 57 | assert cap.isOpened(), 'Cannot capture source' 58 | 59 | print("\nPerforming object detection:") 60 | 61 | # Video feed dimensions 62 | _, frame = cap.read() 63 | v_height, v_width = frame.shape[:2] 64 | 65 | # For a black image 66 | x = y = v_height if v_height > v_width else v_width 67 | 68 | # Putting original image into black image 69 | start_new_i_height = int((y - v_height) / 2) 70 | start_new_i_width = int((x - v_width) / 2) 71 | 72 | # For accommodate results in original frame 73 | mul_constant = x / opt.frame_size 74 | 75 | # for text in output 76 | t_size = cv2.getTextSize(" ", cv2.FONT_HERSHEY_PLAIN, 1, 1)[0] 77 | 78 | frames = fps = 0 79 | start = time.time() 80 | 81 | while _: 82 | 83 | # frame extraction => resizing => [BGR -> RGB] => [[0...255] -> [0...1]] => [[3, 416, 416] -> [416, 416, 3]] 84 | # => [[416, 416, 3] => [416, 416, 3, 1]] => [np_array -> tensor] => [tensor -> variable] 85 | 86 | # frame extraction 87 | _, org_frame = cap.read() 88 | # resizing to [416 x 416] 89 | 90 | # Black image 91 | frame = np.zeros((x, y, 3), np.uint8) 92 | 93 | frame[start_new_i_height: (start_new_i_height + v_height) ,start_new_i_width: (start_new_i_width + v_width) ] = org_frame 94 | 95 | #resizing to [416x 416] 96 | frame = cv2.resize(frame, (opt.frame_size, opt.frame_size)) 97 | # [BGR -> RGB] 98 | frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB) 99 | # [[0...255] -> [0...1]] 100 | frame = np.asarray(frame) / 255 101 | # [[3, 416, 416] -> [416, 416, 3]] 102 | frame = np.transpose(frame, [2, 0, 1]) 103 | # [[416, 416, 3] => [416, 416, 3, 1]] 104 | frame = np.expand_dims(frame, axis=0) 105 | # [np_array -> tensor] 106 | frame = torch.Tensor(frame) 107 | 108 | # plt.imshow(frame[0].permute(1,2,0)) 109 | # plt.show() 110 | 111 | # [tensor -> variable] 112 | frame = Variable(frame.type(Tensor)) 113 | 114 | # Get detections 115 | with torch.no_grad(): 116 | detections = model(frame) 117 | detections = non_max_suppression_output(detections, opt.conf_thres, opt.nms_thres) 118 | 119 | # For each detection in detections 120 | detection = detections[0] 121 | if detection is not None: 122 | 123 | for x1, y1, x2, y2, conf, cls_conf, cls_pred in detection: 124 | 125 | # Accommodate bounding box in original frame 126 | x1 = int(x1 * mul_constant - start_new_i_width) 127 | y1 = int(y1 * mul_constant - start_new_i_height) 128 | x2 = int(x2 * mul_constant - start_new_i_width) 129 | y2 = int(y2 * mul_constant - start_new_i_height) 130 | 131 | # Bounding box making and setting Bounding box title 132 | if (int(cls_pred) == 0): 133 | # WITH_MASK 134 | cv2.rectangle(org_frame, (x1, y1), (x2, y2), (0, 255, 0), 2) 135 | else: 136 | #WITHOUT_MASK 137 | cv2.rectangle(org_frame, (x1, y1), (x2, y2), (0, 0, 255), 2) 138 | 139 | cv2.putText(org_frame, classes[int(cls_pred)]+ ": %.2f" % conf, (x1, y1 + t_size[1]), 140 | cv2.FONT_HERSHEY_PLAIN, 1, 141 | [225, 255, 255], 2) 142 | 143 | # CURRENT TIME SHOWING 144 | now = datetime.now() 145 | current_time = now.strftime("%H:%M:%S") 146 | 147 | # FPS PRINTING 148 | cv2.rectangle(org_frame, (0, 0), (175, 20), (0, 0, 0), -1) 149 | cv2.putText(org_frame, current_time + " FPS : %3.2f" % (fps), (0, t_size[1] + 2), 150 | cv2.FONT_HERSHEY_PLAIN, 1, 151 | [255, 255, 255], 1) 152 | 153 | frames += 1 154 | fps = frames / (time.time() - start) 155 | 156 | cv2.namedWindow('frame', cv2.WINDOW_NORMAL) 157 | cv2.setWindowProperty('frame', cv2.WND_PROP_FULLSCREEN-5, cv2.WINDOW_FULLSCREEN) 158 | 159 | cv2.imshow('frame', org_frame) 160 | if cv2.waitKey(1) & 0xFF == ord('q'): 161 | break 162 | 163 | cap.release() 164 | cv2.destroyAllWindows() 165 | -------------------------------------------------------------------------------- /video_detect.py: -------------------------------------------------------------------------------- 1 | from __future__ import division 2 | 3 | from models import Darknet 4 | from utils.utils import load_classes,non_max_suppression_output 5 | 6 | import argparse 7 | 8 | import time 9 | import cv2 10 | import os 11 | import torch 12 | import numpy as np 13 | from torch.autograd import Variable 14 | 15 | if __name__ == "__main__": 16 | parser = argparse.ArgumentParser() 17 | parser.add_argument("--input_file_path", type=str, default="testing/input/videos/Clip032.mp4", help="path to video file") 18 | parser.add_argument("--model_def", type=str, default="config/yolov3_mask.cfg", help="path to model definition file") 19 | parser.add_argument("--weights_path", type=str, default="checkpoints/yolov3_ckpt_35.pth", help="path to weights file") 20 | parser.add_argument("--class_path", type=str, default="data/mask_dataset.names", help="path to class label file") 21 | parser.add_argument("--conf_thres", type=float, default=0.9, help="object confidence threshold") 22 | parser.add_argument("--nms_thres", type=float, default=0.4, help="iou thresshold for non-maximum suppression") 23 | parser.add_argument("--frame_size", type=int, default=416, help="size of each image dimension") 24 | parser.add_argument("--save_video", type=bool, default=True, help="save output video or not") 25 | parser.add_argument("--output_path", type=str, default="testing/output/videos", help="output video path") 26 | 27 | opt = parser.parse_args() 28 | print(opt) 29 | 30 | # Output directory 31 | os.makedirs(opt.output_path, exist_ok=True) 32 | 33 | # checking for GPU 34 | device = torch.device("cuda" if torch.cuda.is_available() else "cpu") 35 | 36 | # Set up model 37 | model = Darknet(opt.model_def, img_size=opt.frame_size).to(device) 38 | 39 | # loading weights 40 | if opt.weights_path.endswith(".weights"): 41 | model.load_darknet_weights(opt.weights_path) # Load weights 42 | else: 43 | model.load_state_dict(torch.load(opt.weights_path)) # Load checkpoints 44 | 45 | # Set in evaluation mode 46 | model.eval() 47 | 48 | # Extracts class labels from file 49 | classes = load_classes(opt.class_path) 50 | 51 | # ckecking for GPU for Tensor 52 | Tensor = torch.cuda.FloatTensor if torch.cuda.is_available() else torch.FloatTensor 53 | 54 | # camara capture 55 | cap = cv2.VideoCapture(opt.input_file_path) 56 | assert cap.isOpened(), 'Cannot capture source' 57 | 58 | # Video feed dimensions 59 | _, frame = cap.read() 60 | v_height, v_width = frame.shape[:2] 61 | 62 | # print(v_height,v_width) 63 | 64 | # Output saving 65 | if(opt.save_video): 66 | fourcc = cv2.VideoWriter_fourcc(*'MP4V') 67 | 68 | filename = opt.input_file_path.split("/")[-1] 69 | filepath = os.path.join(opt.output_path,filename) 70 | 71 | fps = cap.get(cv2.CAP_PROP_FPS) 72 | out = cv2.VideoWriter(filepath, fourcc, fps, (v_width, v_height)) 73 | 74 | print("\nPerforming object detection:") 75 | 76 | # For a black image 77 | x = y = v_height if v_height > v_width else v_width 78 | 79 | # Putting original image into black image 80 | start_new_i_height = int((y - v_height) / 2) 81 | start_new_i_width = int((x - v_width) / 2) 82 | 83 | # For accommodate results in original frame 84 | mul_constant = x / opt.frame_size 85 | # print(mul_constant) 86 | 87 | # for text in output 88 | t_size = cv2.getTextSize(" ", cv2.FONT_HERSHEY_PLAIN, 1, 1)[0] 89 | 90 | frames = fps = 0 91 | start = time.time() 92 | 93 | while _: 94 | 95 | # frame extraction => resizing => [BGR -> RGB] => [[0...255] -> [0...1]] => [[3, 416, 416] -> [416, 416, 3]] 96 | # => [[416, 416, 3] => [416, 416, 3, 1]] => [np_array -> tensor] => [tensor -> variable] 97 | 98 | # frame extraction 99 | _, org_frame = cap.read() 100 | # resizing to [416 x 416] 101 | 102 | # Black image 103 | frame = np.zeros((x, y, 3), np.uint8) 104 | 105 | frame[start_new_i_height: (start_new_i_height + v_height),start_new_i_width: (start_new_i_width + v_width)] = org_frame 106 | 107 | # resizing to [416x 416] 108 | frame = cv2.resize(frame, (opt.frame_size, opt.frame_size)) 109 | # [BGR -> RGB] 110 | frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB) 111 | # [[0...255] -> [0...1]] 112 | frame = np.asarray(frame) / 255 113 | # [[3, 416, 416] -> [416, 416, 3]] 114 | frame = np.transpose(frame, [2, 0, 1]) 115 | # [[416, 416, 3] => [416, 416, 3, 1]] 116 | frame = np.expand_dims(frame, axis=0) 117 | # [np_array -> tensor] 118 | frame = torch.Tensor(frame) 119 | 120 | # plt.imshow(frame[0].permute(1,2,0)) 121 | # plt.show() 122 | 123 | # [tensor -> variable] 124 | frame = Variable(frame.type(Tensor)) 125 | 126 | # Get detections 127 | with torch.no_grad(): 128 | detections = model(frame) 129 | detections = non_max_suppression_output(detections, opt.conf_thres, opt.nms_thres) 130 | 131 | # For each detection in detections 132 | detection = detections[0] 133 | if detection is not None: 134 | 135 | for x1, y1, x2, y2, conf, cls_conf, cls_pred in detection: 136 | 137 | # Accommodate bounding box in original frame 138 | x1 = int(x1 * mul_constant - start_new_i_width) 139 | y1 = int(y1 * mul_constant - start_new_i_height) 140 | x2 = int(x2 * mul_constant - start_new_i_width) 141 | y2 = int(y2 * mul_constant - start_new_i_height) 142 | 143 | # Bounding box making and setting Bounding box title 144 | if (int(cls_pred) == 0): 145 | # WITH_MASK 146 | cv2.rectangle(org_frame, (x1, y1), (x2, y2), (0, 255, 0), 2) 147 | else: 148 | # WITHOUT_MASK 149 | cv2.rectangle(org_frame, (x1, y1), (x2, y2), (0, 0, 255), 2) 150 | 151 | cv2.putText(org_frame, classes[int(cls_pred)] + ": %.2f" % conf, (x1, y1 + t_size[1] + 4), 152 | cv2.FONT_HERSHEY_PLAIN, 1, 153 | [225, 255, 255], 2) 154 | 155 | 156 | # FPS PRINTING 157 | # cv2.rectangle(org_frame, (0, 0), (175, 20), (0, 0, 0), -1) 158 | # cv2.putText(org_frame,"FPS : %3.2f" % (fps), (0, t_size[1] + 4), 159 | # cv2.FONT_HERSHEY_PLAIN, 1, 160 | # [255, 255, 255], 1) 161 | 162 | frames += 1 163 | fps = frames / (time.time() - start) 164 | 165 | # cv2.namedWindow('frame', cv2.WINDOW_NORMAL) 166 | # cv2.setWindowProperty('frame', cv2.WND_PROP_FULLSCREEN, cv2.WINDOW_FULLSCREEN) 167 | 168 | if (opt.save_video): 169 | out.write(org_frame) 170 | 171 | cv2.imshow('frame', org_frame) 172 | if cv2.waitKey(1) & 0xFF == ord('q'): 173 | break 174 | 175 | if (opt.save_video): 176 | out.release() 177 | 178 | cap.release() 179 | cv2.destroyAllWindows() -------------------------------------------------------------------------------- /config/yolov3_mask.cfg: -------------------------------------------------------------------------------- 1 | [net] 2 | batch=16 3 | subdivisions=1 4 | width=416 5 | height=416 6 | channels=3 7 | momentum=0.9 8 | decay=0.0005 9 | angle=0 10 | saturation = 1.5 11 | exposure = 1.5 12 | hue=.1 13 | 14 | learning_rate=0.0001 15 | burn_in=1000 16 | max_batches = 500200 17 | policy=steps 18 | steps=400000,450000 19 | scales=.1,.1 20 | 21 | [convolutional] 22 | batch_normalize=1 23 | filters=32 24 | size=3 25 | stride=1 26 | pad=1 27 | activation=leaky 28 | 29 | # Downsample 30 | 31 | [convolutional] 32 | batch_normalize=1 33 | filters=64 34 | size=3 35 | stride=2 36 | pad=1 37 | activation=leaky 38 | 39 | [convolutional] 40 | batch_normalize=1 41 | filters=32 42 | size=1 43 | stride=1 44 | pad=1 45 | activation=leaky 46 | 47 | [convolutional] 48 | batch_normalize=1 49 | filters=64 50 | size=3 51 | stride=1 52 | pad=1 53 | activation=leaky 54 | 55 | [shortcut] 56 | from=-3 57 | activation=linear 58 | 59 | # Downsample 60 | 61 | [convolutional] 62 | batch_normalize=1 63 | filters=128 64 | size=3 65 | stride=2 66 | pad=1 67 | activation=leaky 68 | 69 | [convolutional] 70 | batch_normalize=1 71 | filters=64 72 | size=1 73 | stride=1 74 | pad=1 75 | activation=leaky 76 | 77 | [convolutional] 78 | batch_normalize=1 79 | filters=128 80 | size=3 81 | stride=1 82 | pad=1 83 | activation=leaky 84 | 85 | [shortcut] 86 | from=-3 87 | activation=linear 88 | 89 | [convolutional] 90 | batch_normalize=1 91 | filters=64 92 | size=1 93 | stride=1 94 | pad=1 95 | activation=leaky 96 | 97 | [convolutional] 98 | batch_normalize=1 99 | filters=128 100 | size=3 101 | stride=1 102 | pad=1 103 | activation=leaky 104 | 105 | [shortcut] 106 | from=-3 107 | activation=linear 108 | 109 | # Downsample 110 | 111 | [convolutional] 112 | batch_normalize=1 113 | filters=256 114 | size=3 115 | stride=2 116 | pad=1 117 | activation=leaky 118 | 119 | [convolutional] 120 | batch_normalize=1 121 | filters=128 122 | size=1 123 | stride=1 124 | pad=1 125 | activation=leaky 126 | 127 | [convolutional] 128 | batch_normalize=1 129 | filters=256 130 | size=3 131 | stride=1 132 | pad=1 133 | activation=leaky 134 | 135 | [shortcut] 136 | from=-3 137 | activation=linear 138 | 139 | [convolutional] 140 | batch_normalize=1 141 | filters=128 142 | size=1 143 | stride=1 144 | pad=1 145 | activation=leaky 146 | 147 | [convolutional] 148 | batch_normalize=1 149 | filters=256 150 | size=3 151 | stride=1 152 | pad=1 153 | activation=leaky 154 | 155 | [shortcut] 156 | from=-3 157 | activation=linear 158 | 159 | [convolutional] 160 | batch_normalize=1 161 | filters=128 162 | size=1 163 | stride=1 164 | pad=1 165 | activation=leaky 166 | 167 | [convolutional] 168 | batch_normalize=1 169 | filters=256 170 | size=3 171 | stride=1 172 | pad=1 173 | activation=leaky 174 | 175 | [shortcut] 176 | from=-3 177 | activation=linear 178 | 179 | [convolutional] 180 | batch_normalize=1 181 | filters=128 182 | size=1 183 | stride=1 184 | pad=1 185 | activation=leaky 186 | 187 | [convolutional] 188 | batch_normalize=1 189 | filters=256 190 | size=3 191 | stride=1 192 | pad=1 193 | activation=leaky 194 | 195 | [shortcut] 196 | from=-3 197 | activation=linear 198 | 199 | 200 | [convolutional] 201 | batch_normalize=1 202 | filters=128 203 | size=1 204 | stride=1 205 | pad=1 206 | activation=leaky 207 | 208 | [convolutional] 209 | batch_normalize=1 210 | filters=256 211 | size=3 212 | stride=1 213 | pad=1 214 | activation=leaky 215 | 216 | [shortcut] 217 | from=-3 218 | activation=linear 219 | 220 | [convolutional] 221 | batch_normalize=1 222 | filters=128 223 | size=1 224 | stride=1 225 | pad=1 226 | activation=leaky 227 | 228 | [convolutional] 229 | batch_normalize=1 230 | filters=256 231 | size=3 232 | stride=1 233 | pad=1 234 | activation=leaky 235 | 236 | [shortcut] 237 | from=-3 238 | activation=linear 239 | 240 | [convolutional] 241 | batch_normalize=1 242 | filters=128 243 | size=1 244 | stride=1 245 | pad=1 246 | activation=leaky 247 | 248 | [convolutional] 249 | batch_normalize=1 250 | filters=256 251 | size=3 252 | stride=1 253 | pad=1 254 | activation=leaky 255 | 256 | [shortcut] 257 | from=-3 258 | activation=linear 259 | 260 | [convolutional] 261 | batch_normalize=1 262 | filters=128 263 | size=1 264 | stride=1 265 | pad=1 266 | activation=leaky 267 | 268 | [convolutional] 269 | batch_normalize=1 270 | filters=256 271 | size=3 272 | stride=1 273 | pad=1 274 | activation=leaky 275 | 276 | [shortcut] 277 | from=-3 278 | activation=linear 279 | 280 | # Downsample 281 | 282 | [convolutional] 283 | batch_normalize=1 284 | filters=512 285 | size=3 286 | stride=2 287 | pad=1 288 | activation=leaky 289 | 290 | [convolutional] 291 | batch_normalize=1 292 | filters=256 293 | size=1 294 | stride=1 295 | pad=1 296 | activation=leaky 297 | 298 | [convolutional] 299 | batch_normalize=1 300 | filters=512 301 | size=3 302 | stride=1 303 | pad=1 304 | activation=leaky 305 | 306 | [shortcut] 307 | from=-3 308 | activation=linear 309 | 310 | 311 | [convolutional] 312 | batch_normalize=1 313 | filters=256 314 | size=1 315 | stride=1 316 | pad=1 317 | activation=leaky 318 | 319 | [convolutional] 320 | batch_normalize=1 321 | filters=512 322 | size=3 323 | stride=1 324 | pad=1 325 | activation=leaky 326 | 327 | [shortcut] 328 | from=-3 329 | activation=linear 330 | 331 | 332 | [convolutional] 333 | batch_normalize=1 334 | filters=256 335 | size=1 336 | stride=1 337 | pad=1 338 | activation=leaky 339 | 340 | [convolutional] 341 | batch_normalize=1 342 | filters=512 343 | size=3 344 | stride=1 345 | pad=1 346 | activation=leaky 347 | 348 | [shortcut] 349 | from=-3 350 | activation=linear 351 | 352 | 353 | [convolutional] 354 | batch_normalize=1 355 | filters=256 356 | size=1 357 | stride=1 358 | pad=1 359 | activation=leaky 360 | 361 | [convolutional] 362 | batch_normalize=1 363 | filters=512 364 | size=3 365 | stride=1 366 | pad=1 367 | activation=leaky 368 | 369 | [shortcut] 370 | from=-3 371 | activation=linear 372 | 373 | [convolutional] 374 | batch_normalize=1 375 | filters=256 376 | size=1 377 | stride=1 378 | pad=1 379 | activation=leaky 380 | 381 | [convolutional] 382 | batch_normalize=1 383 | filters=512 384 | size=3 385 | stride=1 386 | pad=1 387 | activation=leaky 388 | 389 | [shortcut] 390 | from=-3 391 | activation=linear 392 | 393 | 394 | [convolutional] 395 | batch_normalize=1 396 | filters=256 397 | size=1 398 | stride=1 399 | pad=1 400 | activation=leaky 401 | 402 | [convolutional] 403 | batch_normalize=1 404 | filters=512 405 | size=3 406 | stride=1 407 | pad=1 408 | activation=leaky 409 | 410 | [shortcut] 411 | from=-3 412 | activation=linear 413 | 414 | 415 | [convolutional] 416 | batch_normalize=1 417 | filters=256 418 | size=1 419 | stride=1 420 | pad=1 421 | activation=leaky 422 | 423 | [convolutional] 424 | batch_normalize=1 425 | filters=512 426 | size=3 427 | stride=1 428 | pad=1 429 | activation=leaky 430 | 431 | [shortcut] 432 | from=-3 433 | activation=linear 434 | 435 | [convolutional] 436 | batch_normalize=1 437 | filters=256 438 | size=1 439 | stride=1 440 | pad=1 441 | activation=leaky 442 | 443 | [convolutional] 444 | batch_normalize=1 445 | filters=512 446 | size=3 447 | stride=1 448 | pad=1 449 | activation=leaky 450 | 451 | [shortcut] 452 | from=-3 453 | activation=linear 454 | 455 | # Downsample 456 | 457 | [convolutional] 458 | batch_normalize=1 459 | filters=1024 460 | size=3 461 | stride=2 462 | pad=1 463 | activation=leaky 464 | 465 | [convolutional] 466 | batch_normalize=1 467 | filters=512 468 | size=1 469 | stride=1 470 | pad=1 471 | activation=leaky 472 | 473 | [convolutional] 474 | batch_normalize=1 475 | filters=1024 476 | size=3 477 | stride=1 478 | pad=1 479 | activation=leaky 480 | 481 | [shortcut] 482 | from=-3 483 | activation=linear 484 | 485 | [convolutional] 486 | batch_normalize=1 487 | filters=512 488 | size=1 489 | stride=1 490 | pad=1 491 | activation=leaky 492 | 493 | [convolutional] 494 | batch_normalize=1 495 | filters=1024 496 | size=3 497 | stride=1 498 | pad=1 499 | activation=leaky 500 | 501 | [shortcut] 502 | from=-3 503 | activation=linear 504 | 505 | [convolutional] 506 | batch_normalize=1 507 | filters=512 508 | size=1 509 | stride=1 510 | pad=1 511 | activation=leaky 512 | 513 | [convolutional] 514 | batch_normalize=1 515 | filters=1024 516 | size=3 517 | stride=1 518 | pad=1 519 | activation=leaky 520 | 521 | [shortcut] 522 | from=-3 523 | activation=linear 524 | 525 | [convolutional] 526 | batch_normalize=1 527 | filters=512 528 | size=1 529 | stride=1 530 | pad=1 531 | activation=leaky 532 | 533 | [convolutional] 534 | batch_normalize=1 535 | filters=1024 536 | size=3 537 | stride=1 538 | pad=1 539 | activation=leaky 540 | 541 | [shortcut] 542 | from=-3 543 | activation=linear 544 | 545 | ###################### 546 | 547 | [convolutional] 548 | batch_normalize=1 549 | filters=512 550 | size=1 551 | stride=1 552 | pad=1 553 | activation=leaky 554 | 555 | [convolutional] 556 | batch_normalize=1 557 | size=3 558 | stride=1 559 | pad=1 560 | filters=1024 561 | activation=leaky 562 | 563 | [convolutional] 564 | batch_normalize=1 565 | filters=512 566 | size=1 567 | stride=1 568 | pad=1 569 | activation=leaky 570 | 571 | [convolutional] 572 | batch_normalize=1 573 | size=3 574 | stride=1 575 | pad=1 576 | filters=1024 577 | activation=leaky 578 | 579 | [convolutional] 580 | batch_normalize=1 581 | filters=512 582 | size=1 583 | stride=1 584 | pad=1 585 | activation=leaky 586 | 587 | [convolutional] 588 | batch_normalize=1 589 | size=3 590 | stride=1 591 | pad=1 592 | filters=1024 593 | activation=leaky 594 | 595 | [convolutional] 596 | size=1 597 | stride=1 598 | pad=1 599 | filters=21 600 | activation=linear 601 | 602 | 603 | [yolo] 604 | mask = 6,7,8 605 | anchors = 27, 31, 39, 49, 54, 69, 72, 96, 94, 128, 123, 165, 166, 222, 239, 313, 387, 493 606 | classes=2 607 | num=9 608 | jitter=.3 609 | ignore_thresh = .7 610 | truth_thresh = 1 611 | random=1 612 | 613 | 614 | [route] 615 | layers = -4 616 | 617 | [convolutional] 618 | batch_normalize=1 619 | filters=256 620 | size=1 621 | stride=1 622 | pad=1 623 | activation=leaky 624 | 625 | [upsample] 626 | stride=2 627 | 628 | [route] 629 | layers = -1, 61 630 | 631 | 632 | 633 | [convolutional] 634 | batch_normalize=1 635 | filters=256 636 | size=1 637 | stride=1 638 | pad=1 639 | activation=leaky 640 | 641 | [convolutional] 642 | batch_normalize=1 643 | size=3 644 | stride=1 645 | pad=1 646 | filters=512 647 | activation=leaky 648 | 649 | [convolutional] 650 | batch_normalize=1 651 | filters=256 652 | size=1 653 | stride=1 654 | pad=1 655 | activation=leaky 656 | 657 | [convolutional] 658 | batch_normalize=1 659 | size=3 660 | stride=1 661 | pad=1 662 | filters=512 663 | activation=leaky 664 | 665 | [convolutional] 666 | batch_normalize=1 667 | filters=256 668 | size=1 669 | stride=1 670 | pad=1 671 | activation=leaky 672 | 673 | [convolutional] 674 | batch_normalize=1 675 | size=3 676 | stride=1 677 | pad=1 678 | filters=512 679 | activation=leaky 680 | 681 | [convolutional] 682 | size=1 683 | stride=1 684 | pad=1 685 | filters=21 686 | activation=linear 687 | 688 | 689 | [yolo] 690 | mask = 3,4,5 691 | anchors = 27,31, 39,49, 54,69, 72,96, 94,128, 123,165, 166,222, 239,313, 387,493 692 | classes=2 693 | num=9 694 | jitter=.3 695 | ignore_thresh = .7 696 | truth_thresh = 1 697 | random=1 698 | 699 | 700 | 701 | [route] 702 | layers = -4 703 | 704 | [convolutional] 705 | batch_normalize=1 706 | filters=128 707 | size=1 708 | stride=1 709 | pad=1 710 | activation=leaky 711 | 712 | [upsample] 713 | stride=2 714 | 715 | [route] 716 | layers = -1, 36 717 | 718 | 719 | 720 | [convolutional] 721 | batch_normalize=1 722 | filters=128 723 | size=1 724 | stride=1 725 | pad=1 726 | activation=leaky 727 | 728 | [convolutional] 729 | batch_normalize=1 730 | size=3 731 | stride=1 732 | pad=1 733 | filters=256 734 | activation=leaky 735 | 736 | [convolutional] 737 | batch_normalize=1 738 | filters=128 739 | size=1 740 | stride=1 741 | pad=1 742 | activation=leaky 743 | 744 | [convolutional] 745 | batch_normalize=1 746 | size=3 747 | stride=1 748 | pad=1 749 | filters=256 750 | activation=leaky 751 | 752 | [convolutional] 753 | batch_normalize=1 754 | filters=128 755 | size=1 756 | stride=1 757 | pad=1 758 | activation=leaky 759 | 760 | [convolutional] 761 | batch_normalize=1 762 | size=3 763 | stride=1 764 | pad=1 765 | filters=256 766 | activation=leaky 767 | 768 | [convolutional] 769 | size=1 770 | stride=1 771 | pad=1 772 | filters=21 773 | activation=linear 774 | 775 | 776 | [yolo] 777 | mask = 0,1,2 778 | anchors = 27, 31, 39, 49, 54, 69, 72, 96, 94, 128, 123, 165, 166, 222, 239, 313, 387, 493 779 | classes=2 780 | num=9 781 | jitter=.3 782 | ignore_thresh = .7 783 | truth_thresh = 1 784 | random=1 785 | -------------------------------------------------------------------------------- /models.py: -------------------------------------------------------------------------------- 1 | from __future__ import division 2 | 3 | import torch 4 | import torch.nn as nn 5 | import torch.nn.functional as F 6 | import numpy as np 7 | 8 | from utils.parse_config import parse_model_config 9 | from utils.utils import build_targets, to_cpu 10 | 11 | 12 | def create_modules(module_defs): 13 | """ 14 | Constructs module list of layer blocks from module configuration in module_defs 15 | """ 16 | hyperparams = module_defs.pop(0) 17 | output_filters = [int(hyperparams["channels"])] 18 | module_list = nn.ModuleList() 19 | for module_i, module_def in enumerate(module_defs): 20 | modules = nn.Sequential() 21 | 22 | if module_def["type"] == "convolutional": 23 | bn = int(module_def["batch_normalize"]) 24 | filters = int(module_def["filters"]) 25 | kernel_size = int(module_def["size"]) 26 | pad = (kernel_size - 1) // 2 27 | modules.add_module( 28 | f"conv_{module_i}", 29 | nn.Conv2d( 30 | in_channels=output_filters[-1], 31 | out_channels=filters, 32 | kernel_size=kernel_size, 33 | stride=int(module_def["stride"]), 34 | padding=pad, 35 | bias=not bn, 36 | ), 37 | ) 38 | if bn: 39 | modules.add_module(f"batch_norm_{module_i}", nn.BatchNorm2d(filters, momentum=0.9, eps=1e-5)) 40 | if module_def["activation"] == "leaky": 41 | modules.add_module(f"leaky_{module_i}", nn.LeakyReLU(0.1)) 42 | 43 | elif module_def["type"] == "upsample": 44 | upsample = Upsample(scale_factor=int(module_def["stride"]), mode="nearest") 45 | modules.add_module(f"upsample_{module_i}", upsample) 46 | 47 | elif module_def["type"] == "route": 48 | layers = [int(x) for x in module_def["layers"].split(",")] 49 | filters = sum([output_filters[1:][i] for i in layers]) 50 | modules.add_module(f"route_{module_i}", EmptyLayer()) 51 | 52 | elif module_def["type"] == "shortcut": 53 | filters = output_filters[1:][int(module_def["from"])] 54 | modules.add_module(f"shortcut_{module_i}", EmptyLayer()) 55 | 56 | elif module_def["type"] == "yolo": 57 | anchor_idxs = [int(x) for x in module_def["mask"].split(",")] 58 | # Extract anchors 59 | anchors = [int(x) for x in module_def["anchors"].split(",")] 60 | anchors = [(anchors[i], anchors[i + 1]) for i in range(0, len(anchors), 2)] 61 | anchors = [anchors[i] for i in anchor_idxs] 62 | num_classes = int(module_def["classes"]) 63 | img_size = int(hyperparams["height"]) 64 | # Define detection layer 65 | yolo_layer = YOLOLayer(anchors, num_classes, img_size) 66 | modules.add_module(f"yolo_{module_i}", yolo_layer) 67 | # Register module list and number of output filters 68 | module_list.append(modules) 69 | output_filters.append(filters) 70 | 71 | return hyperparams, module_list 72 | 73 | 74 | class Upsample(nn.Module): 75 | """ nn.Upsample is deprecated """ 76 | 77 | def __init__(self, scale_factor, mode="nearest"): 78 | super(Upsample, self).__init__() 79 | self.scale_factor = scale_factor 80 | self.mode = mode 81 | 82 | def forward(self, x): 83 | x = F.interpolate(x, scale_factor=self.scale_factor, mode=self.mode) 84 | return x 85 | 86 | 87 | class EmptyLayer(nn.Module): 88 | """Placeholder for 'route' and 'shortcut' layers""" 89 | 90 | def __init__(self): 91 | super(EmptyLayer, self).__init__() 92 | 93 | 94 | class YOLOLayer(nn.Module): 95 | """Detection layer""" 96 | 97 | def __init__(self, anchors, num_classes, img_dim=416): 98 | super(YOLOLayer, self).__init__() 99 | self.anchors = anchors 100 | self.num_anchors = len(anchors) 101 | self.num_classes = num_classes 102 | self.ignore_thres = 0.5 103 | self.mse_loss = nn.MSELoss() 104 | self.bce_loss = nn.BCELoss() 105 | self.obj_scale = 1 106 | self.noobj_scale = 100 107 | self.metrics = {} 108 | self.img_dim = img_dim 109 | self.grid_size = 0 # grid size 110 | 111 | def compute_grid_offsets(self, grid_size, cuda=True): 112 | self.grid_size = grid_size 113 | g = self.grid_size 114 | FloatTensor = torch.cuda.FloatTensor if cuda else torch.FloatTensor 115 | self.stride = self.img_dim / self.grid_size 116 | # Calculate offsets for each grid 117 | self.grid_x = torch.arange(g).repeat(g, 1).view([1, 1, g, g]).type(FloatTensor) 118 | self.grid_y = torch.arange(g).repeat(g, 1).t().view([1, 1, g, g]).type(FloatTensor) 119 | self.scaled_anchors = FloatTensor([(a_w / self.stride, a_h / self.stride) for a_w, a_h in self.anchors]) 120 | self.anchor_w = self.scaled_anchors[:, 0:1].view((1, self.num_anchors, 1, 1)) 121 | self.anchor_h = self.scaled_anchors[:, 1:2].view((1, self.num_anchors, 1, 1)) 122 | 123 | def forward(self, x, targets=None, img_dim=None): 124 | 125 | # Tensors for cuda support 126 | FloatTensor = torch.cuda.FloatTensor if x.is_cuda else torch.FloatTensor 127 | LongTensor = torch.cuda.LongTensor if x.is_cuda else torch.LongTensor 128 | ByteTensor = torch.cuda.ByteTensor if x.is_cuda else torch.ByteTensor 129 | 130 | self.img_dim = img_dim 131 | num_samples = x.size(0) 132 | grid_size = x.size(2) 133 | 134 | prediction = ( 135 | x.view(num_samples, self.num_anchors, self.num_classes + 5, grid_size, grid_size) 136 | .permute(0, 1, 3, 4, 2) 137 | .contiguous() 138 | ) 139 | 140 | # Get outputs 141 | x = torch.sigmoid(prediction[..., 0]) # Center x 142 | y = torch.sigmoid(prediction[..., 1]) # Center y 143 | w = prediction[..., 2] # Width 144 | h = prediction[..., 3] # Height 145 | pred_conf = torch.sigmoid(prediction[..., 4]) # Conf 146 | pred_cls = torch.sigmoid(prediction[..., 5:]) # Cls pred. 147 | 148 | # If grid size does not match current we compute new offsets 149 | if grid_size != self.grid_size: 150 | self.compute_grid_offsets(grid_size, cuda=x.is_cuda) 151 | 152 | # Add offset and scale with anchors 153 | pred_boxes = FloatTensor(prediction[..., :4].shape) 154 | pred_boxes[..., 0] = x.data + self.grid_x 155 | pred_boxes[..., 1] = y.data + self.grid_y 156 | pred_boxes[..., 2] = torch.exp(w.data) * self.anchor_w 157 | pred_boxes[..., 3] = torch.exp(h.data) * self.anchor_h 158 | 159 | output = torch.cat( 160 | ( 161 | pred_boxes.view(num_samples, -1, 4) * self.stride, 162 | pred_conf.view(num_samples, -1, 1), 163 | pred_cls.view(num_samples, -1, self.num_classes), 164 | ), 165 | -1, 166 | ) 167 | 168 | if targets is None: 169 | return output, 0 170 | else: 171 | iou_scores, class_mask, obj_mask, noobj_mask, tx, ty, tw, th, tcls, tconf = build_targets( 172 | pred_boxes=pred_boxes, 173 | pred_cls=pred_cls, 174 | target=targets, 175 | anchors=self.scaled_anchors, 176 | ignore_thres=self.ignore_thres, 177 | ) 178 | 179 | # Loss : Mask outputs to ignore non-existing objects (except with conf. loss) 180 | loss_x = self.mse_loss(x[obj_mask], tx[obj_mask]) 181 | loss_y = self.mse_loss(y[obj_mask], ty[obj_mask]) 182 | loss_w = self.mse_loss(w[obj_mask], tw[obj_mask]) 183 | loss_h = self.mse_loss(h[obj_mask], th[obj_mask]) 184 | 185 | # Calculate BCE of objectness score of a bounding box 186 | loss_conf_obj = self.bce_loss(pred_conf[obj_mask], tconf[obj_mask]) 187 | # Calculate BCE of no objectness score of a bounding box 188 | loss_conf_noobj = self.bce_loss(pred_conf[noobj_mask], tconf[noobj_mask]) 189 | 190 | # Scale and Sum above two LOSS 191 | loss_conf = self.obj_scale * loss_conf_obj + self.noobj_scale * loss_conf_noobj 192 | 193 | # Calculate BCE of multi-class predictions of a bounding box 194 | loss_cls = self.bce_loss(pred_cls[obj_mask], tcls[obj_mask]) 195 | 196 | total_loss = loss_x + loss_y + loss_w + loss_h + loss_conf + loss_cls 197 | 198 | return output, total_loss 199 | 200 | 201 | class Darknet(nn.Module): 202 | """YOLOv3 object detection model""" 203 | 204 | def __init__(self, config_path, img_size=416): 205 | super(Darknet, self).__init__() 206 | self.module_defs = parse_model_config(config_path) 207 | self.hyperparams, self.module_list = create_modules(self.module_defs) 208 | self.yolo_layers = [layer[0] for layer in self.module_list if hasattr(layer[0], "metrics")] 209 | self.img_size = img_size 210 | self.seen = 0 211 | self.header_info = np.array([0, 0, 0, self.seen, 0], dtype=np.int32) 212 | 213 | def forward(self, x, targets=None): 214 | img_dim = x.shape[2] 215 | loss = 0 216 | layer_outputs, yolo_outputs = [], [] 217 | for i, (module_def, module) in enumerate(zip(self.module_defs, self.module_list)): 218 | if module_def["type"] in ["convolutional", "upsample"]: 219 | x = module(x) 220 | elif module_def["type"] == "route": 221 | x = torch.cat([layer_outputs[int(layer_i)] for layer_i in module_def["layers"].split(",")], 1) 222 | elif module_def["type"] == "shortcut": 223 | layer_i = int(module_def["from"]) 224 | x = layer_outputs[-1] + layer_outputs[layer_i] 225 | elif module_def["type"] == "yolo": 226 | x, layer_loss = module[0](x, targets, img_dim) 227 | loss += layer_loss 228 | yolo_outputs.append(x) 229 | layer_outputs.append(x) 230 | yolo_outputs = to_cpu(torch.cat(yolo_outputs, 1)) 231 | return yolo_outputs if targets is None else (loss, yolo_outputs) 232 | 233 | def load_darknet_weights(self, weights_path): 234 | """Parses and loads the weights stored in 'weights_path'""" 235 | 236 | # Open the weights file 237 | with open(weights_path, "rb") as f: 238 | header = np.fromfile(f, dtype=np.int32, count=5) # First five are header values 239 | self.header_info = header # Needed to write header when saving weights 240 | self.seen = header[3] # number of images seen during training 241 | weights = np.fromfile(f, dtype=np.float32) # The rest are weights 242 | 243 | # Establish cutoff for loading backbone weights 244 | cutoff = None 245 | if "darknet53.conv.74" in weights_path: 246 | cutoff = 75 247 | 248 | ptr = 0 249 | for i, (module_def, module) in enumerate(zip(self.module_defs, self.module_list)): 250 | if i == cutoff: 251 | break 252 | if module_def["type"] == "convolutional": 253 | conv_layer = module[0] 254 | if module_def["batch_normalize"]: 255 | # Load BN bias, weights, running mean and running variance 256 | bn_layer = module[1] 257 | num_b = bn_layer.bias.numel() # Number of biases 258 | # Bias 259 | bn_b = torch.from_numpy(weights[ptr : ptr + num_b]).view_as(bn_layer.bias) 260 | bn_layer.bias.data.copy_(bn_b) 261 | ptr += num_b 262 | # Weight 263 | bn_w = torch.from_numpy(weights[ptr : ptr + num_b]).view_as(bn_layer.weight) 264 | bn_layer.weight.data.copy_(bn_w) 265 | ptr += num_b 266 | # Running Mean 267 | bn_rm = torch.from_numpy(weights[ptr : ptr + num_b]).view_as(bn_layer.running_mean) 268 | bn_layer.running_mean.data.copy_(bn_rm) 269 | ptr += num_b 270 | # Running Var 271 | bn_rv = torch.from_numpy(weights[ptr : ptr + num_b]).view_as(bn_layer.running_var) 272 | bn_layer.running_var.data.copy_(bn_rv) 273 | ptr += num_b 274 | else: 275 | # Load conv. bias 276 | num_b = conv_layer.bias.numel() 277 | conv_b = torch.from_numpy(weights[ptr : ptr + num_b]).view_as(conv_layer.bias) 278 | conv_layer.bias.data.copy_(conv_b) 279 | ptr += num_b 280 | # Load conv. weights 281 | num_w = conv_layer.weight.numel() 282 | conv_w = torch.from_numpy(weights[ptr : ptr + num_w]).view_as(conv_layer.weight) 283 | conv_layer.weight.data.copy_(conv_w) 284 | ptr += num_w 285 | 286 | def save_darknet_weights(self, path, cutoff=-1): 287 | """ 288 | @:param path - path of the new weights file 289 | @:param cutoff - save layers between 0 and cutoff (cutoff = -1 -> all are saved) 290 | """ 291 | fp = open(path, "wb") 292 | self.header_info[3] = self.seen 293 | self.header_info.tofile(fp) 294 | 295 | # Iterate through layers 296 | for i, (module_def, module) in enumerate(zip(self.module_defs[:cutoff], self.module_list[:cutoff])): 297 | if module_def["type"] == "convolutional": 298 | conv_layer = module[0] 299 | # If batch norm, load bn first 300 | if module_def["batch_normalize"]: 301 | bn_layer = module[1] 302 | bn_layer.bias.data.cpu().numpy().tofile(fp) 303 | bn_layer.weight.data.cpu().numpy().tofile(fp) 304 | bn_layer.running_mean.data.cpu().numpy().tofile(fp) 305 | bn_layer.running_var.data.cpu().numpy().tofile(fp) 306 | # Load conv bias 307 | else: 308 | conv_layer.bias.data.cpu().numpy().tofile(fp) 309 | # Load conv weights 310 | conv_layer.weight.data.cpu().numpy().tofile(fp) 311 | 312 | fp.close() 313 | -------------------------------------------------------------------------------- /utils/utils.py: -------------------------------------------------------------------------------- 1 | from __future__ import division 2 | import tqdm 3 | import torch 4 | import numpy as np 5 | 6 | 7 | def to_cpu(tensor): 8 | return tensor.detach().cpu() 9 | 10 | 11 | def load_classes(path): 12 | """ 13 | Loads class labels at 'path' 14 | """ 15 | fp = open(path, "r") 16 | names = fp.read().split("\n")[:-1] 17 | return names 18 | 19 | 20 | def weights_init_normal(m): 21 | classname = m.__class__.__name__ 22 | if classname.find("Conv") != -1: 23 | torch.nn.init.normal_(m.weight.data, 0.0, 0.02) 24 | elif classname.find("BatchNorm2d") != -1: 25 | torch.nn.init.normal_(m.weight.data, 1.0, 0.02) 26 | torch.nn.init.constant_(m.bias.data, 0.0) 27 | 28 | def xywh2xyxy(x): 29 | y = x.new(x.shape) 30 | y[..., 0] = x[..., 0] - x[..., 2] / 2 31 | y[..., 1] = x[..., 1] - x[..., 3] / 2 32 | y[..., 2] = x[..., 0] + x[..., 2] / 2 33 | y[..., 3] = x[..., 1] + x[..., 3] / 2 34 | return y 35 | 36 | 37 | def ap_per_class(tp, conf, pred_cls, target_cls): 38 | """ Compute the average precision, given the recall and precision curves. 39 | Source: https://github.com/rafaelpadilla/Object-Detection-Metrics. 40 | # Arguments 41 | tp: True positives (list). 42 | conf: Objectness value from 0-1 (list). 43 | pred_cls: Predicted object classes (list). 44 | target_cls: True object classes (list). 45 | # Returns 46 | The average precision as computed in py-faster-rcnn. 47 | """ 48 | 49 | # Sort by objectness 50 | i = np.argsort(-conf) 51 | tp, conf, pred_cls = tp[i], conf[i], pred_cls[i] 52 | 53 | # Find unique classes 54 | unique_classes = np.unique(target_cls) 55 | 56 | # Create Precision-Recall curve and compute AP for each class 57 | ap, p, r = [], [], [] 58 | for c in tqdm.tqdm(unique_classes, desc="Computing AP"): 59 | i = pred_cls == c 60 | n_gt = (target_cls == c).sum() # Number of ground truth objects 61 | n_p = i.sum() # Number of predicted objects 62 | 63 | if n_p == 0 and n_gt == 0: 64 | continue 65 | elif n_p == 0 or n_gt == 0: 66 | ap.append(0) 67 | r.append(0) 68 | p.append(0) 69 | else: 70 | # Accumulate FPs and TPs 71 | fpc = (1 - tp[i]).cumsum() 72 | tpc = (tp[i]).cumsum() 73 | 74 | # Recall 75 | recall_curve = tpc / (n_gt + 1e-16) 76 | r.append(recall_curve[-1]) 77 | 78 | # Precision 79 | precision_curve = tpc / (tpc + fpc) 80 | p.append(precision_curve[-1]) 81 | 82 | # AP from recall-precision curve 83 | ap.append(compute_ap(recall_curve, precision_curve)) 84 | 85 | # Compute F1 score (harmonic mean of precision and recall) 86 | p, r, ap = np.array(p), np.array(r), np.array(ap) 87 | f1 = 2 * p * r / (p + r + 1e-16) 88 | 89 | return p, r, ap, f1, unique_classes.astype("int32") 90 | 91 | 92 | def compute_ap(recall, precision): 93 | """ Compute the average precision, given the recall and precision curves. 94 | Code originally from https://github.com/rbgirshick/py-faster-rcnn. 95 | 96 | # Arguments 97 | recall: The recall curve (list). 98 | precision: The precision curve (list). 99 | # Returns 100 | The average precision as computed in py-faster-rcnn. 101 | """ 102 | # correct AP calculation 103 | # first append sentinel values at the end 104 | mrec = np.concatenate(([0.0], recall, [1.0])) 105 | mpre = np.concatenate(([0.0], precision, [0.0])) 106 | 107 | # compute the precision envelope 108 | for i in range(mpre.size - 1, 0, -1): 109 | mpre[i - 1] = np.maximum(mpre[i - 1], mpre[i]) 110 | 111 | # to calculate area under PR curve, look for points 112 | # where X axis (recall) changes value 113 | i = np.where(mrec[1:] != mrec[:-1])[0] 114 | 115 | # and sum (\Delta recall) * prec 116 | ap = np.sum((mrec[i + 1] - mrec[i]) * mpre[i + 1]) 117 | return ap 118 | 119 | 120 | def get_batch_statistics(outputs, targets, iou_threshold): 121 | """ Compute true positives, predicted scores and predicted labels per sample """ 122 | batch_metrics = [] 123 | for sample_i in range(len(outputs)): 124 | 125 | if outputs[sample_i] is None: 126 | continue 127 | 128 | output = outputs[sample_i] 129 | pred_boxes = output[:, :4] 130 | pred_scores = output[:, 4] 131 | pred_labels = output[:, -1] 132 | 133 | true_positives = np.zeros(pred_boxes.shape[0]) 134 | 135 | annotations = targets[targets[:, 0] == sample_i][:, 1:] 136 | target_labels = annotations[:, 0] if len(annotations) else [] 137 | if len(annotations): 138 | detected_boxes = [] 139 | target_boxes = annotations[:, 1:] 140 | 141 | for pred_i, (pred_box, pred_label) in enumerate(zip(pred_boxes, pred_labels)): 142 | 143 | # If targets are found break 144 | if len(detected_boxes) == len(annotations): 145 | break 146 | 147 | # Ignore if label is not one of the target labels 148 | if pred_label not in target_labels: 149 | continue 150 | 151 | iou, box_index = bbox_iou(pred_box.unsqueeze(0), target_boxes).max(0) 152 | if iou >= iou_threshold and box_index not in detected_boxes: 153 | true_positives[pred_i] = 1 154 | detected_boxes += [box_index] 155 | batch_metrics.append([true_positives, pred_scores, pred_labels]) 156 | return batch_metrics 157 | 158 | 159 | def bbox_wh_iou(wh1, wh2): 160 | wh2 = wh2.t() 161 | w1, h1 = wh1[0], wh1[1] 162 | w2, h2 = wh2[0], wh2[1] 163 | inter_area = torch.min(w1, w2) * torch.min(h1, h2) 164 | union_area = (w1 * h1 + 1e-16) + w2 * h2 - inter_area 165 | return inter_area / union_area 166 | 167 | 168 | def bbox_iou(box1, box2, x1y1x2y2=True): 169 | """ 170 | Returns the IoU of two bounding boxes 171 | """ 172 | if not x1y1x2y2: 173 | # Transform from center and width to exact coordinates 174 | b1_x1, b1_x2 = box1[:, 0] - box1[:, 2] / 2, box1[:, 0] + box1[:, 2] / 2 175 | b1_y1, b1_y2 = box1[:, 1] - box1[:, 3] / 2, box1[:, 1] + box1[:, 3] / 2 176 | b2_x1, b2_x2 = box2[:, 0] - box2[:, 2] / 2, box2[:, 0] + box2[:, 2] / 2 177 | b2_y1, b2_y2 = box2[:, 1] - box2[:, 3] / 2, box2[:, 1] + box2[:, 3] / 2 178 | else: 179 | # Get the coordinates of bounding boxes 180 | b1_x1, b1_y1, b1_x2, b1_y2 = box1[:, 0], box1[:, 1], box1[:, 2], box1[:, 3] 181 | b2_x1, b2_y1, b2_x2, b2_y2 = box2[:, 0], box2[:, 1], box2[:, 2], box2[:, 3] 182 | 183 | # get the corrdinates of the intersection rectangle 184 | inter_rect_x1 = torch.max(b1_x1, b2_x1) 185 | inter_rect_y1 = torch.max(b1_y1, b2_y1) 186 | inter_rect_x2 = torch.min(b1_x2, b2_x2) 187 | inter_rect_y2 = torch.min(b1_y2, b2_y2) 188 | # Intersection area 189 | inter_area = torch.clamp(inter_rect_x2 - inter_rect_x1 + 1, min=0) * torch.clamp( 190 | inter_rect_y2 - inter_rect_y1 + 1, min=0 191 | ) 192 | # Union Area 193 | b1_area = (b1_x2 - b1_x1 + 1) * (b1_y2 - b1_y1 + 1) 194 | b2_area = (b2_x2 - b2_x1 + 1) * (b2_y2 - b2_y1 + 1) 195 | 196 | iou = inter_area / (b1_area + b2_area - inter_area + 1e-16) 197 | 198 | return iou 199 | 200 | def non_max_suppression_output(prediction, conf_thres=0.5, nms_thres=0.4): 201 | """ 202 | Removes detections with lower object confidence score than 'conf_thres' and performs 203 | Non-Maximum Suppression to further filter detections. 204 | Returns detections with shape: 205 | (x1, y1, x2, y2, object_conf, class_score, class_pred) """ 206 | 207 | # From (center x, center y, width, height) to (x1, y1, x2, y2) 208 | prediction[..., :4] = xywh2xyxy(prediction[..., :4]) 209 | output = [None for _ in range(len(prediction))] 210 | 211 | for image_i, image_pred in enumerate(prediction): 212 | 213 | # Filter out confidence scores below threshold 214 | image_pred = image_pred[image_pred[:, 4] >= conf_thres] 215 | 216 | # If none are remaining => process next image 217 | if not image_pred.size(0): 218 | continue 219 | 220 | # Object confidence times class confidence 221 | score = image_pred[:, 4] * image_pred[:, 5:].max(1)[0] 222 | 223 | # Sort by it 224 | image_pred = image_pred[(-score).argsort()] 225 | class_confs, class_preds = image_pred[:, 5:].max(1, keepdim=True) 226 | 227 | detections = torch.cat((image_pred[:, :5], class_confs.float(), class_preds.float()),1) 228 | 229 | # print(detections) 230 | 231 | # Perform non-maximum suppression 232 | keep_boxes = [] 233 | while detections.size(0): 234 | large_overlap = bbox_iou(detections[0, :4].unsqueeze(0), detections[:, :4]) > nms_thres 235 | 236 | # print("------------------------{0}------------------------".format(detections.size(0))) 237 | # print(torch.nonzero(large_overlap).size()[0]) 238 | 239 | label_match = detections[0, -1] == detections[:, -1] 240 | 241 | # Indices of boxes with lower confidence scores, large IOUs and matching labels 242 | invalid = large_overlap & label_match 243 | # print(torch.nonzero(invalid).size()[0]) 244 | 245 | weights = detections[large_overlap, 4:5] 246 | # Merge overlapping bboxes by order of confidence 247 | 248 | detections[0, :4] = (weights * detections[large_overlap, :4]).sum(0) / weights.sum() 249 | 250 | if not (torch.nonzero(large_overlap).size()[0] == torch.nonzero(invalid).size()[0]): 251 | detections[0,-1] = 0.0 252 | 253 | keep_boxes += [detections[0]] 254 | detections = detections[~large_overlap] 255 | if keep_boxes: 256 | output[image_i] = torch.stack(keep_boxes) 257 | 258 | return output 259 | 260 | 261 | def non_max_suppression(prediction, conf_thres=0.5, nms_thres=0.4): 262 | """ 263 | Removes detections with lower object confidence score than 'conf_thres' and performs 264 | Non-Maximum Suppression to further filter detections. 265 | Returns detections with shape: 266 | (x1, y1, x2, y2, object_conf, class_score, class_pred) 267 | """ 268 | 269 | # From (center x, center y, width, height) to (x1, y1, x2, y2) 270 | prediction[..., :4] = xywh2xyxy(prediction[..., :4]) 271 | output = [None for _ in range(len(prediction))] 272 | 273 | for image_i, image_pred in enumerate(prediction): 274 | 275 | # Filter out confidence scores below threshold 276 | image_pred = image_pred[image_pred[:, 4] >= conf_thres] 277 | 278 | # If none are remaining => process next image 279 | if not image_pred.size(0): 280 | continue 281 | 282 | # Object confidence times class confidence 283 | score = image_pred[:, 4] * image_pred[:, 5:].max(1)[0] 284 | 285 | # Sort by it 286 | image_pred = image_pred[(-score).argsort()] 287 | class_confs, class_preds = image_pred[:, 5:].max(1, keepdim=True) 288 | 289 | detections = torch.cat((image_pred[:, :5], class_confs.float(), class_preds.float()),1) 290 | 291 | # print(detections) 292 | 293 | # Perform non-maximum suppression 294 | keep_boxes = [] 295 | while detections.size(0): 296 | large_overlap = bbox_iou(detections[0, :4].unsqueeze(0), detections[:, :4]) > nms_thres 297 | 298 | label_match = detections[0, -1] == detections[:, -1] 299 | 300 | # Indices of boxes with lower confidence scores, large IOUs and matching labels 301 | invalid = large_overlap & label_match 302 | weights = detections[invalid, 4:5] 303 | # Merge overlapping bboxes by order of confidence 304 | detections[0, :4] = (weights * detections[invalid, :4]).sum(0) / weights.sum() 305 | keep_boxes += [detections[0]] 306 | detections = detections[~invalid] 307 | if keep_boxes: 308 | output[image_i] = torch.stack(keep_boxes) 309 | 310 | return output 311 | 312 | 313 | def build_targets(pred_boxes, pred_cls, target, anchors, ignore_thres): 314 | 315 | ByteTensor = torch.cuda.ByteTensor if pred_boxes.is_cuda else torch.ByteTensor 316 | FloatTensor = torch.cuda.FloatTensor if pred_boxes.is_cuda else torch.FloatTensor 317 | 318 | nB = pred_boxes.size(0) 319 | nA = pred_boxes.size(1) 320 | nC = pred_cls.size(-1) 321 | nG = pred_boxes.size(2) 322 | 323 | # Output tensors 324 | obj_mask = ByteTensor(nB, nA, nG, nG).fill_(0) 325 | noobj_mask = ByteTensor(nB, nA, nG, nG).fill_(1) 326 | class_mask = FloatTensor(nB, nA, nG, nG).fill_(0) 327 | iou_scores = FloatTensor(nB, nA, nG, nG).fill_(0) 328 | tx = FloatTensor(nB, nA, nG, nG).fill_(0) 329 | ty = FloatTensor(nB, nA, nG, nG).fill_(0) 330 | tw = FloatTensor(nB, nA, nG, nG).fill_(0) 331 | th = FloatTensor(nB, nA, nG, nG).fill_(0) 332 | tcls = FloatTensor(nB, nA, nG, nG, nC).fill_(0) 333 | 334 | # Convert to position relative to box 335 | target_boxes = target[:, 2:6] * nG 336 | gxy = target_boxes[:, :2] 337 | gwh = target_boxes[:, 2:] 338 | # Get anchors with best iou 339 | ious = torch.stack([bbox_wh_iou(anchor, gwh) for anchor in anchors]) 340 | best_ious, best_n = ious.max(0) 341 | # Separate target values 342 | b, target_labels = target[:, :2].long().t() 343 | gx, gy = gxy.t() 344 | gw, gh = gwh.t() 345 | gi, gj = gxy.long().t() 346 | 347 | ########## 348 | gi[gi < 0] = 0 349 | gj[gj < 0] = 0 350 | gi[gi > nG - 1] = nG - 1 351 | gj[gj > nG - 1] = nG - 1 352 | ################### 353 | 354 | # Set masks 355 | obj_mask[b, best_n, gj, gi] = 1 356 | noobj_mask[b, best_n, gj, gi] = 0 357 | 358 | # Set noobj mask to zero where iou exceeds ignore threshold 359 | for i, anchor_ious in enumerate(ious.t()): 360 | noobj_mask[b[i], anchor_ious > ignore_thres, gj[i], gi[i]] = 0 361 | 362 | # Coordinates 363 | tx[b, best_n, gj, gi] = gx - gx.floor() 364 | ty[b, best_n, gj, gi] = gy - gy.floor() 365 | # Width and height 366 | tw[b, best_n, gj, gi] = torch.log(gw / anchors[best_n][:, 0] + 1e-16) 367 | th[b, best_n, gj, gi] = torch.log(gh / anchors[best_n][:, 1] + 1e-16) 368 | # One-hot encoding of label 369 | tcls[b, best_n, gj, gi, target_labels] = 1 370 | # Compute label correctness and iou at best anchor 371 | class_mask[b, best_n, gj, gi] = (pred_cls[b, best_n, gj, gi].argmax(-1) == target_labels).float() 372 | iou_scores[b, best_n, gj, gi] = bbox_iou(pred_boxes[b, best_n, gj, gi], target_boxes, x1y1x2y2=False) 373 | 374 | tconf = obj_mask.float() 375 | return iou_scores, class_mask, obj_mask, noobj_mask, tx, ty, tw, th, tcls, tconf 376 | -------------------------------------------------------------------------------- /data/mask_dataset_validate.txt: -------------------------------------------------------------------------------- 1 | ..\Traing_Data\images\CLIP064F00219.jpg 2 | ..\Traing_Data\images\CLIP072F00326.jpg 3 | ..\Traing_Data\images\CLIP003906.jpg 4 | ..\Traing_Data\images\CLIP073F00904.jpg 5 | ..\Traing_Data\images\CLIP072F00111.jpg 6 | ..\Traing_Data\images\CLIP000559.jpg 7 | ..\Traing_Data\images\CLIP010909.jpg 8 | ..\Traing_Data\images\CLIP066F00010.jpg 9 | ..\Traing_Data\images\CLIP001818.jpg 10 | ..\Traing_Data\images\CLIP072F01020.jpg 11 | ..\Traing_Data\images\CLIP074F00555.jpg 12 | ..\Traing_Data\images\CLIP012374.jpg 13 | ..\Traing_Data\images\CLIP000143.jpg 14 | ..\Traing_Data\images\CLIP002521.jpg 15 | ..\Traing_Data\images\CLIP001583.jpg 16 | ..\Traing_Data\images\CLIP080F00044.jpg 17 | ..\Traing_Data\images\CLIP010320.jpg 18 | ..\Traing_Data\images\CLIP011185.jpg 19 | ..\Traing_Data\images\CLIP068F00084.jpg 20 | ..\Traing_Data\images\CLIP000921.jpg 21 | ..\Traing_Data\images\CLIP009781.jpg 22 | ..\Traing_Data\images\CLIP073F00595.jpg 23 | ..\Traing_Data\images\CLIP008035.jpg 24 | ..\Traing_Data\images\2034.jpg 25 | ..\Traing_Data\images\CLIP011496.jpg 26 | ..\Traing_Data\images\CLIP006643.jpg 27 | ..\Traing_Data\images\CLIP001829.jpg 28 | ..\Traing_Data\images\CLIP004690.jpg 29 | ..\Traing_Data\images\CLIP071F00319.jpg 30 | ..\Traing_Data\images\CLIP003903.jpg 31 | ..\Traing_Data\images\CLIP004229.jpg 32 | ..\Traing_Data\images\CLIP075F00053.jpg 33 | ..\Traing_Data\images\CLIP072F00071.jpg 34 | ..\Traing_Data\images\CLIP000393.jpg 35 | ..\Traing_Data\images\CLIP006027.jpg 36 | ..\Traing_Data\images\CLIP010304.jpg 37 | ..\Traing_Data\images\1631.jpg 38 | ..\Traing_Data\images\CLIP012518.jpg 39 | ..\Traing_Data\images\CLIP008518.jpg 40 | ..\Traing_Data\images\CLIP064F00333.jpg 41 | ..\Traing_Data\images\CLIP006770.jpg 42 | ..\Traing_Data\images\1635.jpg 43 | ..\Traing_Data\images\CLIP006938.jpg 44 | ..\Traing_Data\images\CLIP064F00339.jpg 45 | ..\Traing_Data\images\CLIP072F00774.jpg 46 | ..\Traing_Data\images\CLIP074F01071.jpg 47 | ..\Traing_Data\images\CLIP008943.jpg 48 | ..\Traing_Data\images\CLIP071F00129.jpg 49 | ..\Traing_Data\images\CLIP012949.jpg 50 | ..\Traing_Data\images\CLIP011029.jpg 51 | ..\Traing_Data\images\CLIP073F01212.jpg 52 | ..\Traing_Data\images\CLIP001406.jpg 53 | ..\Traing_Data\images\CLIP014057.jpg 54 | ..\Traing_Data\images\CLIP072F00851.jpg 55 | ..\Traing_Data\images\CLIP009145.jpg 56 | ..\Traing_Data\images\CLIP072F00092.jpg 57 | ..\Traing_Data\images\CLIP069F00012.jpg 58 | ..\Traing_Data\images\CLIP072F00598.jpg 59 | ..\Traing_Data\images\CLIP073F01018.jpg 60 | ..\Traing_Data\images\CLIP073F00656.jpg 61 | ..\Traing_Data\images\3125.jpg 62 | ..\Traing_Data\images\CLIP079F00006.jpg 63 | ..\Traing_Data\images\CLIP010673.jpg 64 | ..\Traing_Data\images\CLIP080F00043.jpg 65 | ..\Traing_Data\images\CLIP012622.jpg 66 | ..\Traing_Data\images\CLIP000573.jpg 67 | ..\Traing_Data\images\CLIP070F00046.jpg 68 | ..\Traing_Data\images\CLIP074F00266.jpg 69 | ..\Traing_Data\images\CLIP002516.jpg 70 | ..\Traing_Data\images\CLIP064F00076.jpg 71 | ..\Traing_Data\images\CLIP072F00017.jpg 72 | ..\Traing_Data\images\CLIP002021.jpg 73 | ..\Traing_Data\images\CLIP074F00001.jpg 74 | ..\Traing_Data\images\4176.jpg 75 | ..\Traing_Data\images\4060.jpg 76 | ..\Traing_Data\images\CLIP013048.jpg 77 | ..\Traing_Data\images\CLIP077F01557.jpg 78 | ..\Traing_Data\images\CLIP065F00005.jpg 79 | ..\Traing_Data\images\CLIP070F00068.jpg 80 | ..\Traing_Data\images\CLIP070F00067.jpg 81 | ..\Traing_Data\images\CLIP011599.jpg 82 | ..\Traing_Data\images\CLIP002265.jpg 83 | ..\Traing_Data\images\1662.jpg 84 | ..\Traing_Data\images\CLIP072F00960.jpg 85 | ..\Traing_Data\images\CLIP008418.jpg 86 | ..\Traing_Data\images\CLIP001938.jpg 87 | ..\Traing_Data\images\CLIP075F00463.jpg 88 | ..\Traing_Data\images\CLIP072F00193.jpg 89 | ..\Traing_Data\images\CLIP010815.jpg 90 | ..\Traing_Data\images\CLIP077F01402.jpg 91 | ..\Traing_Data\images\CLIP069F00064.jpg 92 | ..\Traing_Data\images\2016.jpg 93 | ..\Traing_Data\images\CLIP074F00452.jpg 94 | ..\Traing_Data\images\CLIP071F00172.jpg 95 | ..\Traing_Data\images\CLIP073F00619.jpg 96 | ..\Traing_Data\images\CLIP005348.jpg 97 | ..\Traing_Data\images\CLIP077F00498.jpg 98 | ..\Traing_Data\images\CLIP071F00318.jpg 99 | ..\Traing_Data\images\CLIP003377.jpg 100 | ..\Traing_Data\images\CLIP071F00296.jpg 101 | ..\Traing_Data\images\CLIP072F00023.jpg 102 | ..\Traing_Data\images\CLIP072F00226.jpg 103 | ..\Traing_Data\images\CLIP073F00779.jpg 104 | ..\Traing_Data\images\CLIP011776.jpg 105 | ..\Traing_Data\images\CLIP011681.jpg 106 | ..\Traing_Data\images\CLIP069F00030.jpg 107 | ..\Traing_Data\images\CLIP072F00192.jpg 108 | ..\Traing_Data\images\CLIP011295.jpg 109 | ..\Traing_Data\images\CLIP071F00108.jpg 110 | ..\Traing_Data\images\CLIP000814.jpg 111 | ..\Traing_Data\images\647.jpg 112 | ..\Traing_Data\images\CLIP072F00247.jpg 113 | ..\Traing_Data\images\CLIP004791.jpg 114 | ..\Traing_Data\images\CLIP007854.jpg 115 | ..\Traing_Data\images\CLIP000241.jpg 116 | ..\Traing_Data\images\CLIP074F00613.jpg 117 | ..\Traing_Data\images\CLIP068F00076.jpg 118 | ..\Traing_Data\images\CLIP076F00009.jpg 119 | ..\Traing_Data\images\CLIP000165.jpg 120 | ..\Traing_Data\images\CLIP075F00056.jpg 121 | ..\Traing_Data\images\CLIP075F00998.jpg 122 | ..\Traing_Data\images\CLIP073F00877.jpg 123 | ..\Traing_Data\images\CLIP005462.jpg 124 | ..\Traing_Data\images\CLIP072F00861.jpg 125 | ..\Traing_Data\images\1013.jpg 126 | ..\Traing_Data\images\CLIP074F00727.jpg 127 | ..\Traing_Data\images\CLIP010839.jpg 128 | ..\Traing_Data\images\3006.jpg 129 | ..\Traing_Data\images\CLIP003298.jpg 130 | ..\Traing_Data\images\CLIP075F00468.jpg 131 | ..\Traing_Data\images\CLIP072F01018.jpg 132 | ..\Traing_Data\images\1598.jpg 133 | ..\Traing_Data\images\3214.jpg 134 | ..\Traing_Data\images\CLIP073F00642.jpg 135 | ..\Traing_Data\images\CLIP072F00874.jpg 136 | ..\Traing_Data\images\CLIP072F00899.jpg 137 | ..\Traing_Data\images\CLIP073F00584.jpg 138 | ..\Traing_Data\images\CLIP001896.jpg 139 | ..\Traing_Data\images\CLIP079F00516.jpg 140 | ..\Traing_Data\images\CLIP010597.jpg 141 | ..\Traing_Data\images\CLIP002595.jpg 142 | ..\Traing_Data\images\3365.jpg 143 | ..\Traing_Data\images\CLIP012529.jpg 144 | ..\Traing_Data\images\CLIP079F00458.jpg 145 | ..\Traing_Data\images\CLIP072F00253.jpg 146 | ..\Traing_Data\images\CLIP069F00083.jpg 147 | ..\Traing_Data\images\CLIP005615.jpg 148 | ..\Traing_Data\images\CLIP073F00784.jpg 149 | ..\Traing_Data\images\CLIP078F00973.jpg 150 | ..\Traing_Data\images\CLIP072F00176.jpg 151 | ..\Traing_Data\images\CLIP003420.jpg 152 | ..\Traing_Data\images\CLIP002873.jpg 153 | ..\Traing_Data\images\CLIP072F01014.jpg 154 | ..\Traing_Data\images\CLIP077F00303.jpg 155 | ..\Traing_Data\images\CLIP073F00644.jpg 156 | ..\Traing_Data\images\CLIP005749.jpg 157 | ..\Traing_Data\images\CLIP066F00012.jpg 158 | ..\Traing_Data\images\CLIP072F00952.jpg 159 | ..\Traing_Data\images\CLIP075F01002.jpg 160 | ..\Traing_Data\images\362.jpg 161 | ..\Traing_Data\images\CLIP010194.jpg 162 | ..\Traing_Data\images\CLIP013084.jpg 163 | ..\Traing_Data\images\CLIP073F00973.jpg 164 | ..\Traing_Data\images\CLIP004619.jpg 165 | ..\Traing_Data\images\CLIP073F01146.jpg 166 | ..\Traing_Data\images\CLIP072F00937.jpg 167 | ..\Traing_Data\images\CLIP064F00349.jpg 168 | ..\Traing_Data\images\CLIP000609.jpg 169 | ..\Traing_Data\images\CLIP068F00104.jpg 170 | ..\Traing_Data\images\CLIP008573.jpg 171 | ..\Traing_Data\images\CLIP007589.jpg 172 | ..\Traing_Data\images\CLIP005354.jpg 173 | ..\Traing_Data\images\3390.jpg 174 | ..\Traing_Data\images\CLIP010083.jpg 175 | ..\Traing_Data\images\CLIP067F00024.jpg 176 | ..\Traing_Data\images\CLIP074F01068.jpg 177 | ..\Traing_Data\images\CLIP004069.jpg 178 | ..\Traing_Data\images\CLIP071F00195.jpg 179 | ..\Traing_Data\images\CLIP066F00136.jpg 180 | ..\Traing_Data\images\CLIP072F00094.jpg 181 | ..\Traing_Data\images\CLIP074F00478.jpg 182 | ..\Traing_Data\images\CLIP000846.jpg 183 | ..\Traing_Data\images\CLIP001731.jpg 184 | ..\Traing_Data\images\CLIP069F00103.jpg 185 | ..\Traing_Data\images\CLIP013405.jpg 186 | ..\Traing_Data\images\CLIP007775.jpg 187 | ..\Traing_Data\images\CLIP068F00010.jpg 188 | ..\Traing_Data\images\CLIP010564.jpg 189 | ..\Traing_Data\images\CLIP001127.jpg 190 | ..\Traing_Data\images\CLIP074F00632.jpg 191 | ..\Traing_Data\images\CLIP008230.jpg 192 | ..\Traing_Data\images\CLIP003087.jpg 193 | ..\Traing_Data\images\CLIP075F00429.jpg 194 | ..\Traing_Data\images\CLIP008350.jpg 195 | ..\Traing_Data\images\CLIP009335.jpg 196 | ..\Traing_Data\images\CLIP075F01624.jpg 197 | ..\Traing_Data\images\CLIP001828.jpg 198 | ..\Traing_Data\images\CLIP001814.jpg 199 | ..\Traing_Data\images\CLIP073F00638.jpg 200 | ..\Traing_Data\images\2042.jpg 201 | ..\Traing_Data\images\CLIP073F01259.jpg 202 | ..\Traing_Data\images\CLIP069F00073.jpg 203 | ..\Traing_Data\images\CLIP074F00497.jpg 204 | ..\Traing_Data\images\CLIP073F00612.jpg 205 | ..\Traing_Data\images\CLIP013924.jpg 206 | ..\Traing_Data\images\CLIP071F00187.jpg 207 | ..\Traing_Data\images\CLIP001720.jpg 208 | ..\Traing_Data\images\1088.jpg 209 | ..\Traing_Data\images\CLIP071F00259.jpg 210 | ..\Traing_Data\images\CLIP011672.jpg 211 | ..\Traing_Data\images\CLIP071F00299.jpg 212 | ..\Traing_Data\images\CLIP068F00099.jpg 213 | ..\Traing_Data\images\CLIP073F00893.jpg 214 | ..\Traing_Data\images\CLIP002826.jpg 215 | ..\Traing_Data\images\CLIP074F00171.jpg 216 | ..\Traing_Data\images\CLIP004309.jpg 217 | ..\Traing_Data\images\CLIP068F00034.jpg 218 | ..\Traing_Data\images\CLIP011504.jpg 219 | ..\Traing_Data\images\CLIP071F00285.jpg 220 | ..\Traing_Data\images\CLIP005374.jpg 221 | ..\Traing_Data\images\CLIP005997.jpg 222 | ..\Traing_Data\images\CLIP001046.jpg 223 | ..\Traing_Data\images\CLIP007772.jpg 224 | ..\Traing_Data\images\CLIP007679.jpg 225 | ..\Traing_Data\images\CLIP075F01209.jpg 226 | ..\Traing_Data\images\CLIP074F00285.jpg 227 | ..\Traing_Data\images\CLIP014076.jpg 228 | ..\Traing_Data\images\CLIP072F00074.jpg 229 | ..\Traing_Data\images\CLIP068F00009.jpg 230 | ..\Traing_Data\images\CLIP072F00157.jpg 231 | ..\Traing_Data\images\CLIP074F00550.jpg 232 | ..\Traing_Data\images\CLIP005162.jpg 233 | ..\Traing_Data\images\CLIP071F00239.jpg 234 | ..\Traing_Data\images\2168.jpg 235 | ..\Traing_Data\images\CLIP068F00058.jpg 236 | ..\Traing_Data\images\CLIP010622.jpg 237 | ..\Traing_Data\images\CLIP068F00051.jpg 238 | ..\Traing_Data\images\CLIP071F00293.jpg 239 | ..\Traing_Data\images\CLIP074F00611.jpg 240 | ..\Traing_Data\images\CLIP000852.jpg 241 | ..\Traing_Data\images\CLIP012056.jpg 242 | ..\Traing_Data\images\CLIP072F00245.jpg 243 | ..\Traing_Data\images\3922.jpg 244 | ..\Traing_Data\images\1580.jpg 245 | ..\Traing_Data\images\2973.jpg 246 | ..\Traing_Data\images\CLIP074F00299.jpg 247 | ..\Traing_Data\images\CLIP008465.jpg 248 | ..\Traing_Data\images\CLIP005949.jpg 249 | ..\Traing_Data\images\2851.jpg 250 | ..\Traing_Data\images\CLIP008568.jpg 251 | ..\Traing_Data\images\CLIP064F00006.jpg 252 | ..\Traing_Data\images\1944.jpg 253 | ..\Traing_Data\images\CLIP066F00036.jpg 254 | ..\Traing_Data\images\CLIP008349.jpg 255 | ..\Traing_Data\images\CLIP004223.jpg 256 | ..\Traing_Data\images\CLIP072F00594.jpg 257 | ..\Traing_Data\images\CLIP079F00693.jpg 258 | ..\Traing_Data\images\CLIP001696.jpg 259 | ..\Traing_Data\images\CLIP073F00894.jpg 260 | ..\Traing_Data\images\1486.jpg 261 | ..\Traing_Data\images\CLIP007564.jpg 262 | ..\Traing_Data\images\CLIP075F00606.jpg 263 | ..\Traing_Data\images\CLIP013217.jpg 264 | ..\Traing_Data\images\CLIP080F00491.jpg 265 | ..\Traing_Data\images\CLIP064F00093.jpg 266 | ..\Traing_Data\images\CLIP004751.jpg 267 | ..\Traing_Data\images\CLIP077F01560.jpg 268 | ..\Traing_Data\images\CLIP072F00550.jpg 269 | ..\Traing_Data\images\CLIP075F02108.jpg 270 | ..\Traing_Data\images\CLIP013652.jpg 271 | ..\Traing_Data\images\CLIP003443.jpg 272 | ..\Traing_Data\images\CLIP073F00622.jpg 273 | ..\Traing_Data\images\CLIP008387.jpg 274 | ..\Traing_Data\images\CLIP074F00634.jpg 275 | ..\Traing_Data\images\CLIP003138.jpg 276 | ..\Traing_Data\images\CLIP007639.jpg 277 | ..\Traing_Data\images\CLIP066F00059.jpg 278 | ..\Traing_Data\images\CLIP009879.jpg 279 | ..\Traing_Data\images\CLIP008989.jpg 280 | ..\Traing_Data\images\CLIP074F00468.jpg 281 | ..\Traing_Data\images\CLIP075F00505.jpg 282 | ..\Traing_Data\images\CLIP066F00090.jpg 283 | ..\Traing_Data\images\CLIP077F01867.jpg 284 | ..\Traing_Data\images\CLIP064F00013.jpg 285 | ..\Traing_Data\images\CLIP008621.jpg 286 | ..\Traing_Data\images\CLIP071F00169.jpg 287 | ..\Traing_Data\images\CLIP009152.jpg 288 | ..\Traing_Data\images\CLIP072F00248.jpg 289 | ..\Traing_Data\images\CLIP002153.jpg 290 | ..\Traing_Data\images\CLIP071F00245.jpg 291 | ..\Traing_Data\images\CLIP075F01314.jpg 292 | ..\Traing_Data\images\3746.jpg 293 | ..\Traing_Data\images\CLIP072F00003.jpg 294 | ..\Traing_Data\images\CLIP006816.jpg 295 | ..\Traing_Data\images\4127.jpg 296 | ..\Traing_Data\images\CLIP071F00273.jpg 297 | ..\Traing_Data\images\CLIP071F00208.jpg 298 | ..\Traing_Data\images\3532.jpg 299 | ..\Traing_Data\images\CLIP067F00015.jpg 300 | ..\Traing_Data\images\CLIP072F00294.jpg 301 | ..\Traing_Data\images\CLIP071F00196.jpg 302 | ..\Traing_Data\images\116.jpg 303 | ..\Traing_Data\images\CLIP064F00017.jpg 304 | ..\Traing_Data\images\CLIP076F00533.jpg 305 | ..\Traing_Data\images\1653.jpg 306 | ..\Traing_Data\images\2048.jpg 307 | ..\Traing_Data\images\CLIP005380.jpg 308 | ..\Traing_Data\images\CLIP071F00317.jpg 309 | ..\Traing_Data\images\CLIP006414.jpg 310 | ..\Traing_Data\images\CLIP071F00138.jpg 311 | ..\Traing_Data\images\CLIP003026.jpg 312 | ..\Traing_Data\images\CLIP072F01016.jpg 313 | ..\Traing_Data\images\CLIP070F00016.jpg 314 | ..\Traing_Data\images\1219.jpg 315 | ..\Traing_Data\images\CLIP005867.jpg 316 | ..\Traing_Data\images\CLIP069F00019.jpg 317 | ..\Traing_Data\images\CLIP003887.jpg 318 | ..\Traing_Data\images\CLIP000219.jpg 319 | ..\Traing_Data\images\1141.jpg 320 | ..\Traing_Data\images\CLIP077F01561.jpg 321 | ..\Traing_Data\images\CLIP007749.jpg 322 | ..\Traing_Data\images\CLIP012604.jpg 323 | ..\Traing_Data\images\CLIP010542.jpg 324 | ..\Traing_Data\images\1830.jpg 325 | ..\Traing_Data\images\CLIP007526.jpg 326 | ..\Traing_Data\images\CLIP009219.jpg 327 | ..\Traing_Data\images\CLIP000386.jpg 328 | ..\Traing_Data\images\CLIP071F00000.jpg 329 | ..\Traing_Data\images\CLIP010384.jpg 330 | ..\Traing_Data\images\CLIP073F00896.jpg 331 | ..\Traing_Data\images\CLIP069F00090.jpg 332 | ..\Traing_Data\images\CLIP004744.jpg 333 | ..\Traing_Data\images\CLIP071F00152.jpg 334 | ..\Traing_Data\images\CLIP072F00696.jpg 335 | ..\Traing_Data\images\CLIP071F00086.jpg 336 | ..\Traing_Data\images\CLIP003985.jpg 337 | ..\Traing_Data\images\CLIP003467.jpg 338 | ..\Traing_Data\images\CLIP007401.jpg 339 | ..\Traing_Data\images\CLIP071F00231.jpg 340 | ..\Traing_Data\images\CLIP009878.jpg 341 | ..\Traing_Data\images\CLIP072F01003.jpg 342 | ..\Traing_Data\images\CLIP064F00331.jpg 343 | ..\Traing_Data\images\CLIP004405.jpg 344 | ..\Traing_Data\images\1681.jpg 345 | ..\Traing_Data\images\CLIP007073.jpg 346 | ..\Traing_Data\images\CLIP000726.jpg 347 | ..\Traing_Data\images\CLIP069F00061.jpg 348 | ..\Traing_Data\images\CLIP072F00259.jpg 349 | ..\Traing_Data\images\CLIP010856.jpg 350 | ..\Traing_Data\images\CLIP071F00260.jpg 351 | ..\Traing_Data\images\CLIP074F00214.jpg 352 | ..\Traing_Data\images\CLIP072F00876.jpg 353 | ..\Traing_Data\images\CLIP073F00652.jpg 354 | ..\Traing_Data\images\2701.jpg 355 | ..\Traing_Data\images\CLIP075F00052.jpg 356 | ..\Traing_Data\images\672.jpg 357 | ..\Traing_Data\images\CLIP002787.jpg 358 | ..\Traing_Data\images\786.jpg 359 | ..\Traing_Data\images\CLIP073F00641.jpg 360 | ..\Traing_Data\images\CLIP079F00808.jpg 361 | ..\Traing_Data\images\CLIP008606.jpg 362 | ..\Traing_Data\images\CLIP072F01013.jpg 363 | ..\Traing_Data\images\CLIP079F00071.jpg 364 | ..\Traing_Data\images\3872.jpg 365 | ..\Traing_Data\images\CLIP007896.jpg 366 | ..\Traing_Data\images\CLIP074F00274.jpg 367 | ..\Traing_Data\images\CLIP009627.jpg 368 | ..\Traing_Data\images\CLIP075F00456.jpg 369 | ..\Traing_Data\images\CLIP068F00079.jpg 370 | ..\Traing_Data\images\CLIP011853.jpg 371 | ..\Traing_Data\images\CLIP079F00001.jpg 372 | ..\Traing_Data\images\CLIP007117.jpg 373 | ..\Traing_Data\images\CLIP071F00305.jpg 374 | ..\Traing_Data\images\CLIP007488.jpg 375 | ..\Traing_Data\images\CLIP075F00546.jpg 376 | ..\Traing_Data\images\CLIP000403.jpg 377 | ..\Traing_Data\images\CLIP074F00408.jpg 378 | ..\Traing_Data\images\CLIP080F00455.jpg 379 | ..\Traing_Data\images\3649.jpg 380 | ..\Traing_Data\images\CLIP004578.jpg 381 | ..\Traing_Data\images\CLIP005456.jpg 382 | ..\Traing_Data\images\CLIP075F00750.jpg 383 | ..\Traing_Data\images\CLIP009232.jpg 384 | ..\Traing_Data\images\1857.jpg 385 | ..\Traing_Data\images\CLIP002461.jpg 386 | ..\Traing_Data\images\CLIP069F00151.jpg 387 | ..\Traing_Data\images\CLIP009277.jpg 388 | ..\Traing_Data\images\CLIP000348.jpg 389 | ..\Traing_Data\images\CLIP074F00207.jpg 390 | ..\Traing_Data\images\262.jpg 391 | ..\Traing_Data\images\CLIP001041.jpg 392 | ..\Traing_Data\images\CLIP010966.jpg 393 | ..\Traing_Data\images\CLIP004060.jpg 394 | ..\Traing_Data\images\CLIP008201.jpg 395 | ..\Traing_Data\images\CLIP010690.jpg 396 | ..\Traing_Data\images\CLIP004006.jpg 397 | ..\Traing_Data\images\CLIP064F00320.jpg 398 | ..\Traing_Data\images\CLIP072F00311.jpg 399 | ..\Traing_Data\images\3428.jpg 400 | ..\Traing_Data\images\3910.jpg 401 | ..\Traing_Data\images\CLIP072F00180.jpg 402 | ..\Traing_Data\images\CLIP012620.jpg 403 | ..\Traing_Data\images\CLIP072F00515.jpg 404 | ..\Traing_Data\images\CLIP002555.jpg 405 | ..\Traing_Data\images\CLIP006088.jpg 406 | ..\Traing_Data\images\CLIP006411.jpg 407 | ..\Traing_Data\images\77.jpg 408 | ..\Traing_Data\images\CLIP008374.jpg 409 | ..\Traing_Data\images\CLIP072F00252.jpg 410 | ..\Traing_Data\images\CLIP000189.jpg 411 | ..\Traing_Data\images\1430.jpg 412 | ..\Traing_Data\images\2848.jpg 413 | ..\Traing_Data\images\CLIP071F00146.jpg 414 | ..\Traing_Data\images\CLIP072F00160.jpg 415 | ..\Traing_Data\images\4121.jpg 416 | ..\Traing_Data\images\CLIP006149.jpg 417 | ..\Traing_Data\images\CLIP003932.jpg 418 | ..\Traing_Data\images\CLIP004096.jpg 419 | ..\Traing_Data\images\CLIP076F00320.jpg 420 | ..\Traing_Data\images\CLIP080F00029.jpg 421 | ..\Traing_Data\images\CLIP069F00046.jpg 422 | ..\Traing_Data\images\2802.jpg 423 | ..\Traing_Data\images\CLIP065F00006.jpg 424 | ..\Traing_Data\images\CLIP006485.jpg 425 | ..\Traing_Data\images\CLIP071F00147.jpg 426 | ..\Traing_Data\images\CLIP064F00074.jpg 427 | ..\Traing_Data\images\CLIP072F00700.jpg 428 | ..\Traing_Data\images\CLIP007484.jpg 429 | ..\Traing_Data\images\CLIP075F00759.jpg 430 | ..\Traing_Data\images\CLIP005108.jpg 431 | ..\Traing_Data\images\3197.jpg 432 | ..\Traing_Data\images\CLIP001351.jpg 433 | ..\Traing_Data\images\CLIP071F00012.jpg 434 | ..\Traing_Data\images\CLIP068F00073.jpg 435 | ..\Traing_Data\images\CLIP073F00680.jpg 436 | ..\Traing_Data\images\CLIP011733.jpg 437 | ..\Traing_Data\images\CLIP010701.jpg 438 | ..\Traing_Data\images\CLIP001180.jpg 439 | ..\Traing_Data\images\CLIP074F00357.jpg 440 | ..\Traing_Data\images\CLIP064F00325.jpg 441 | ..\Traing_Data\images\CLIP012507.jpg 442 | ..\Traing_Data\images\CLIP004601.jpg 443 | ..\Traing_Data\images\CLIP013678.jpg 444 | ..\Traing_Data\images\CLIP073F00411.jpg 445 | ..\Traing_Data\images\CLIP075F00517.jpg 446 | ..\Traing_Data\images\CLIP072F00543.jpg 447 | ..\Traing_Data\images\CLIP007507.jpg 448 | ..\Traing_Data\images\73.jpg 449 | ..\Traing_Data\images\CLIP072F00776.jpg 450 | ..\Traing_Data\images\CLIP000747.jpg 451 | ..\Traing_Data\images\777.jpg 452 | ..\Traing_Data\images\CLIP069F00153.jpg 453 | ..\Traing_Data\images\CLIP073F00971.jpg 454 | ..\Traing_Data\images\CLIP077F00308.jpg 455 | ..\Traing_Data\images\CLIP066F00092.jpg 456 | ..\Traing_Data\images\CLIP007180.jpg 457 | ..\Traing_Data\images\CLIP075F00445.jpg 458 | ..\Traing_Data\images\CLIP009541.jpg 459 | ..\Traing_Data\images\CLIP002738.jpg 460 | ..\Traing_Data\images\CLIP010728.jpg 461 | ..\Traing_Data\images\CLIP075F00430.jpg 462 | ..\Traing_Data\images\CLIP075F00588.jpg 463 | ..\Traing_Data\images\CLIP013979.jpg 464 | ..\Traing_Data\images\CLIP072F00856.jpg 465 | ..\Traing_Data\images\CLIP073F01051.jpg 466 | ..\Traing_Data\images\CLIP071F00022.jpg 467 | ..\Traing_Data\images\CLIP013876.jpg 468 | ..\Traing_Data\images\CLIP073F00598.jpg 469 | ..\Traing_Data\images\CLIP074F00489.jpg 470 | ..\Traing_Data\images\CLIP010776.jpg 471 | ..\Traing_Data\images\CLIP002252.jpg 472 | ..\Traing_Data\images\CLIP073F00672.jpg 473 | ..\Traing_Data\images\CLIP073F00024.jpg 474 | ..\Traing_Data\images\CLIP072F00907.jpg 475 | ..\Traing_Data\images\CLIP072F00168.jpg 476 | ..\Traing_Data\images\CLIP000614.jpg 477 | ..\Traing_Data\images\CLIP003031.jpg 478 | ..\Traing_Data\images\CLIP073F01148.jpg 479 | ..\Traing_Data\images\CLIP065F00015.jpg 480 | ..\Traing_Data\images\1222.jpg 481 | ..\Traing_Data\images\CLIP075F00743.jpg 482 | ..\Traing_Data\images\CLIP075F00518.jpg 483 | ..\Traing_Data\images\CLIP072F00940.jpg 484 | ..\Traing_Data\images\4056.jpg 485 | ..\Traing_Data\images\CLIP003698.jpg 486 | ..\Traing_Data\images\1702.jpg 487 | ..\Traing_Data\images\CLIP075F00424.jpg 488 | ..\Traing_Data\images\CLIP072F00559.jpg 489 | ..\Traing_Data\images\CLIP073F01312.jpg 490 | ..\Traing_Data\images\1676.jpg 491 | ..\Traing_Data\images\CLIP073F00903.jpg 492 | ..\Traing_Data\images\CLIP003732.jpg 493 | ..\Traing_Data\images\CLIP070F00043.jpg 494 | ..\Traing_Data\images\CLIP071F00001.jpg 495 | ..\Traing_Data\images\CLIP003123.jpg 496 | ..\Traing_Data\images\CLIP073F00025.jpg 497 | ..\Traing_Data\images\CLIP002868.jpg 498 | ..\Traing_Data\images\CLIP072F00331.jpg 499 | ..\Traing_Data\images\CLIP075F00675.jpg 500 | ..\Traing_Data\images\CLIP071F00303.jpg 501 | ..\Traing_Data\images\CLIP073F00618.jpg 502 | ..\Traing_Data\images\CLIP069F00159.jpg 503 | ..\Traing_Data\images\CLIP003301.jpg 504 | ..\Traing_Data\images\CLIP076F00291.jpg 505 | ..\Traing_Data\images\CLIP075F00611.jpg 506 | ..\Traing_Data\images\CLIP012904.jpg 507 | ..\Traing_Data\images\CLIP075F00607.jpg 508 | ..\Traing_Data\images\CLIP079F00518.jpg 509 | ..\Traing_Data\images\CLIP068F00052.jpg 510 | ..\Traing_Data\images\CLIP008089.jpg 511 | ..\Traing_Data\images\CLIP064F00158.jpg 512 | ..\Traing_Data\images\CLIP001499.jpg 513 | ..\Traing_Data\images\CLIP064F00031.jpg 514 | ..\Traing_Data\images\CLIP002598.jpg 515 | ..\Traing_Data\images\CLIP010920.jpg 516 | ..\Traing_Data\images\CLIP073F00602.jpg 517 | ..\Traing_Data\images\CLIP011156.jpg 518 | ..\Traing_Data\images\CLIP073F01149.jpg 519 | ..\Traing_Data\images\CLIP008802.jpg 520 | ..\Traing_Data\images\101.jpg 521 | ..\Traing_Data\images\CLIP074F00371.jpg 522 | ..\Traing_Data\images\CLIP072F00250.jpg 523 | ..\Traing_Data\images\CLIP011095.jpg 524 | ..\Traing_Data\images\CLIP072F00009.jpg 525 | ..\Traing_Data\images\CLIP012420.jpg 526 | ..\Traing_Data\images\CLIP077F00478.jpg 527 | ..\Traing_Data\images\CLIP005974.jpg 528 | ..\Traing_Data\images\CLIP072F00249.jpg 529 | ..\Traing_Data\images\CLIP012486.jpg 530 | ..\Traing_Data\images\3972.jpg 531 | ..\Traing_Data\images\CLIP072F00902.jpg 532 | ..\Traing_Data\images\CLIP010819.jpg 533 | ..\Traing_Data\images\CLIP073F00433.jpg 534 | ..\Traing_Data\images\CLIP064F00073.jpg 535 | ..\Traing_Data\images\CLIP072F00692.jpg 536 | ..\Traing_Data\images\CLIP070F00018.jpg 537 | ..\Traing_Data\images\CLIP077F00811.jpg 538 | ..\Traing_Data\images\CLIP009679.jpg 539 | ..\Traing_Data\images\CLIP064F00304.jpg 540 | ..\Traing_Data\images\CLIP074F00289.jpg 541 | ..\Traing_Data\images\CLIP072F00038.jpg 542 | ..\Traing_Data\images\CLIP073F00783.jpg 543 | ..\Traing_Data\images\CLIP007313.jpg 544 | ..\Traing_Data\images\CLIP004928.jpg 545 | ..\Traing_Data\images\CLIP000781.jpg 546 | ..\Traing_Data\images\CLIP009774.jpg 547 | ..\Traing_Data\images\CLIP064F00213.jpg 548 | ..\Traing_Data\images\CLIP008695.jpg 549 | ..\Traing_Data\images\CLIP067F00017.jpg 550 | ..\Traing_Data\images\1746.jpg 551 | ..\Traing_Data\images\CLIP011790.jpg 552 | ..\Traing_Data\images\CLIP072F00027.jpg 553 | ..\Traing_Data\images\4097.jpg 554 | ..\Traing_Data\images\CLIP080F00032.jpg 555 | ..\Traing_Data\images\CLIP073F00609.jpg 556 | ..\Traing_Data\images\CLIP000595.jpg 557 | ..\Traing_Data\images\142.jpg 558 | ..\Traing_Data\images\1904.jpg 559 | ..\Traing_Data\images\111.jpg 560 | ..\Traing_Data\images\CLIP072F00950.jpg 561 | ..\Traing_Data\images\CLIP000924.jpg 562 | ..\Traing_Data\images\CLIP076F00322.jpg 563 | ..\Traing_Data\images\CLIP005301.jpg 564 | ..\Traing_Data\images\CLIP012887.jpg 565 | ..\Traing_Data\images\CLIP003817.jpg 566 | ..\Traing_Data\images\CLIP068F00068.jpg 567 | ..\Traing_Data\images\CLIP072F00244.jpg 568 | ..\Traing_Data\images\CLIP070F00051.jpg 569 | ..\Traing_Data\images\791.jpg 570 | ..\Traing_Data\images\CLIP078F03120.jpg 571 | ..\Traing_Data\images\2694.jpg 572 | ..\Traing_Data\images\CLIP072F01005.jpg 573 | ..\Traing_Data\images\CLIP077F00635.jpg 574 | ..\Traing_Data\images\CLIP068F00002.jpg 575 | ..\Traing_Data\images\CLIP005293.jpg 576 | ..\Traing_Data\images\CLIP072F00112.jpg 577 | ..\Traing_Data\images\CLIP068F00017.jpg 578 | ..\Traing_Data\images\CLIP066F00042.jpg 579 | ..\Traing_Data\images\CLIP012550.jpg 580 | ..\Traing_Data\images\CLIP007394.jpg 581 | ..\Traing_Data\images\CLIP003085.jpg 582 | ..\Traing_Data\images\CLIP003886.jpg 583 | ..\Traing_Data\images\CLIP074F00733.jpg 584 | ..\Traing_Data\images\CLIP001428.jpg 585 | ..\Traing_Data\images\3628.jpg 586 | ..\Traing_Data\images\CLIP072F01073.jpg 587 | ..\Traing_Data\images\CLIP072F00131.jpg 588 | ..\Traing_Data\images\CLIP065F00014.jpg 589 | ..\Traing_Data\images\CLIP078F03123.jpg 590 | ..\Traing_Data\images\CLIP006237.jpg 591 | ..\Traing_Data\images\CLIP079F00550.jpg 592 | ..\Traing_Data\images\CLIP072F00911.jpg 593 | ..\Traing_Data\images\CLIP071F00144.jpg 594 | ..\Traing_Data\images\CLIP004347.jpg 595 | ..\Traing_Data\images\CLIP005152.jpg 596 | ..\Traing_Data\images\3713.jpg 597 | ..\Traing_Data\images\CLIP000106.jpg 598 | ..\Traing_Data\images\CLIP007615.jpg 599 | ..\Traing_Data\images\CLIP069F00100.jpg 600 | ..\Traing_Data\images\CLIP003620.jpg 601 | ..\Traing_Data\images\CLIP064F00212.jpg 602 | ..\Traing_Data\images\2906.jpg 603 | ..\Traing_Data\images\CLIP008547.jpg 604 | ..\Traing_Data\images\2072.jpg 605 | ..\Traing_Data\images\CLIP073F00590.jpg 606 | ..\Traing_Data\images\CLIP074F00361.jpg 607 | ..\Traing_Data\images\CLIP008059.jpg 608 | ..\Traing_Data\images\CLIP002140.jpg 609 | ..\Traing_Data\images\CLIP012748.jpg 610 | ..\Traing_Data\images\CLIP003444.jpg 611 | ..\Traing_Data\images\CLIP012713.jpg 612 | ..\Traing_Data\images\CLIP002550.jpg 613 | ..\Traing_Data\images\1454.jpg 614 | ..\Traing_Data\images\CLIP072F00134.jpg 615 | ..\Traing_Data\images\CLIP073F00673.jpg 616 | ..\Traing_Data\images\CLIP012706.jpg 617 | ..\Traing_Data\images\CLIP066F00089.jpg 618 | ..\Traing_Data\images\CLIP074F01015.jpg 619 | ..\Traing_Data\images\CLIP073F00658.jpg 620 | ..\Traing_Data\images\CLIP066F00065.jpg 621 | ..\Traing_Data\images\CLIP004856.jpg 622 | ..\Traing_Data\images\CLIP013567.jpg 623 | ..\Traing_Data\images\CLIP071F00096.jpg 624 | ..\Traing_Data\images\CLIP009169.jpg 625 | ..\Traing_Data\images\CLIP008485.jpg 626 | ..\Traing_Data\images\CLIP072F00561.jpg 627 | ..\Traing_Data\images\CLIP066F00067.jpg 628 | ..\Traing_Data\images\CLIP004400.jpg 629 | ..\Traing_Data\images\3480.jpg 630 | ..\Traing_Data\images\CLIP077F01493.jpg 631 | ..\Traing_Data\images\CLIP008255.jpg 632 | ..\Traing_Data\images\1763.jpg 633 | ..\Traing_Data\images\922.jpg 634 | ..\Traing_Data\images\CLIP001710.jpg 635 | ..\Traing_Data\images\CLIP075F00541.jpg 636 | ..\Traing_Data\images\CLIP073F00417.jpg 637 | ..\Traing_Data\images\CLIP004157.jpg 638 | ..\Traing_Data\images\CLIP068F00089.jpg 639 | ..\Traing_Data\images\CLIP073F01017.jpg 640 | ..\Traing_Data\images\CLIP075F00670.jpg 641 | ..\Traing_Data\images\CLIP077F01873.jpg 642 | ..\Traing_Data\images\CLIP071F00223.jpg 643 | ..\Traing_Data\images\CLIP013179.jpg 644 | ..\Traing_Data\images\CLIP009860.jpg 645 | ..\Traing_Data\images\CLIP079F00660.jpg 646 | ..\Traing_Data\images\CLIP073F00623.jpg 647 | ..\Traing_Data\images\CLIP073F00620.jpg 648 | ..\Traing_Data\images\CLIP075F00529.jpg 649 | ..\Traing_Data\images\CLIP004067.jpg 650 | ..\Traing_Data\images\CLIP069F00148.jpg 651 | ..\Traing_Data\images\CLIP069F00024.jpg 652 | ..\Traing_Data\images\CLIP002602.jpg 653 | ..\Traing_Data\images\CLIP064F00175.jpg 654 | ..\Traing_Data\images\3023.jpg 655 | ..\Traing_Data\images\CLIP013846.jpg 656 | ..\Traing_Data\images\2333.jpg 657 | ..\Traing_Data\images\CLIP077F00515.jpg 658 | ..\Traing_Data\images\CLIP008143.jpg 659 | ..\Traing_Data\images\CLIP064F00233.jpg 660 | ..\Traing_Data\images\CLIP064F00251.jpg 661 | ..\Traing_Data\images\CLIP009848.jpg 662 | ..\Traing_Data\images\CLIP071F00237.jpg 663 | ..\Traing_Data\images\CLIP064F00147.jpg 664 | ..\Traing_Data\images\CLIP006055.jpg 665 | ..\Traing_Data\images\CLIP006204.jpg 666 | ..\Traing_Data\images\CLIP069F00104.jpg 667 | ..\Traing_Data\images\CLIP072F00934.jpg 668 | ..\Traing_Data\images\CLIP073F00410.jpg 669 | ..\Traing_Data\images\CLIP005930.jpg 670 | ..\Traing_Data\images\CLIP071F00083.jpg 671 | ..\Traing_Data\images\CLIP013194.jpg 672 | ..\Traing_Data\images\3594.jpg 673 | ..\Traing_Data\images\CLIP075F00997.jpg 674 | ..\Traing_Data\images\CLIP005285.jpg 675 | ..\Traing_Data\images\CLIP008075.jpg 676 | ..\Traing_Data\images\CLIP072F00046.jpg 677 | ..\Traing_Data\images\CLIP072F00152.jpg 678 | ..\Traing_Data\images\CLIP072F00068.jpg 679 | ..\Traing_Data\images\CLIP001819.jpg 680 | ..\Traing_Data\images\CLIP075F00586.jpg 681 | ..\Traing_Data\images\CLIP073F00408.jpg 682 | ..\Traing_Data\images\CLIP071F00232.jpg 683 | ..\Traing_Data\images\CLIP074F00481.jpg 684 | ..\Traing_Data\images\CLIP071F00302.jpg 685 | ..\Traing_Data\images\CLIP072F00579.jpg 686 | ..\Traing_Data\images\CLIP013367.jpg 687 | ..\Traing_Data\images\CLIP008348.jpg 688 | ..\Traing_Data\images\CLIP074F00277.jpg 689 | ..\Traing_Data\images\CLIP072F00011.jpg 690 | ..\Traing_Data\images\CLIP010248.jpg 691 | ..\Traing_Data\images\432.jpg 692 | ..\Traing_Data\images\CLIP013236.jpg 693 | ..\Traing_Data\images\CLIP068F00016.jpg 694 | ..\Traing_Data\images\CLIP007794.jpg 695 | ..\Traing_Data\images\CLIP007911.jpg 696 | ..\Traing_Data\images\CLIP064F00402.jpg 697 | ..\Traing_Data\images\CLIP001443.jpg 698 | ..\Traing_Data\images\CLIP074F00470.jpg 699 | ..\Traing_Data\images\CLIP069F00150.jpg 700 | ..\Traing_Data\images\CLIP012333.jpg 701 | ..\Traing_Data\images\CLIP073F00616.jpg 702 | ..\Traing_Data\images\CLIP011530.jpg 703 | ..\Traing_Data\images\CLIP072F00610.jpg 704 | ..\Traing_Data\images\CLIP003541.jpg 705 | ..\Traing_Data\images\4049.jpg 706 | ..\Traing_Data\images\CLIP073F00788.jpg 707 | ..\Traing_Data\images\2025.jpg 708 | ..\Traing_Data\images\CLIP003215.jpg 709 | ..\Traing_Data\images\CLIP003627.jpg 710 | ..\Traing_Data\images\CLIP009191.jpg 711 | ..\Traing_Data\images\CLIP072F00171.jpg 712 | ..\Traing_Data\images\4089.jpg 713 | ..\Traing_Data\images\CLIP005991.jpg 714 | ..\Traing_Data\images\CLIP075F00749.jpg 715 | ..\Traing_Data\images\CLIP074F00537.jpg 716 | ..\Traing_Data\images\CLIP001417.jpg 717 | ..\Traing_Data\images\CLIP064F00387.jpg 718 | ..\Traing_Data\images\CLIP011930.jpg 719 | ..\Traing_Data\images\CLIP008849.jpg 720 | ..\Traing_Data\images\CLIP074F00351.jpg 721 | ..\Traing_Data\images\CLIP003543.jpg 722 | ..\Traing_Data\images\CLIP009224.jpg 723 | ..\Traing_Data\images\CLIP075F00472.jpg 724 | ..\Traing_Data\images\CLIP010189.jpg 725 | ..\Traing_Data\images\CLIP004957.jpg 726 | ..\Traing_Data\images\CLIP074F00455.jpg 727 | ..\Traing_Data\images\CLIP075F00523.jpg 728 | ..\Traing_Data\images\CLIP012176.jpg 729 | ..\Traing_Data\images\CLIP072F00237.jpg 730 | ..\Traing_Data\images\CLIP073F00614.jpg 731 | ..\Traing_Data\images\CLIP075F00542.jpg 732 | ..\Traing_Data\images\CLIP072F00233.jpg 733 | ..\Traing_Data\images\CLIP068F00020.jpg 734 | ..\Traing_Data\images\3733.jpg 735 | ..\Traing_Data\images\CLIP013314.jpg 736 | ..\Traing_Data\images\CLIP002634.jpg 737 | ..\Traing_Data\images\1408.jpg 738 | ..\Traing_Data\images\CLIP073F00036.jpg 739 | ..\Traing_Data\images\CLIP064F00167.jpg 740 | ..\Traing_Data\images\114.jpg 741 | ..\Traing_Data\images\CLIP010340.jpg 742 | ..\Traing_Data\images\2603.jpg 743 | ..\Traing_Data\images\CLIP072F00916.jpg 744 | ..\Traing_Data\images\CLIP075F00741.jpg 745 | ..\Traing_Data\images\2515.jpg 746 | ..\Traing_Data\images\CLIP003770.jpg 747 | ..\Traing_Data\images\757.jpg 748 | ..\Traing_Data\images\CLIP003262.jpg 749 | ..\Traing_Data\images\CLIP072F00519.jpg 750 | ..\Traing_Data\images\CLIP009112.jpg 751 | ..\Traing_Data\images\CLIP000124.jpg 752 | ..\Traing_Data\images\CLIP072F00565.jpg 753 | ..\Traing_Data\images\CLIP001427.jpg 754 | ..\Traing_Data\images\CLIP008373.jpg 755 | ..\Traing_Data\images\CLIP076F00042.jpg 756 | ..\Traing_Data\images\CLIP010024.jpg 757 | ..\Traing_Data\images\CLIP072F00531.jpg 758 | ..\Traing_Data\images\CLIP073F01019.jpg 759 | ..\Traing_Data\images\CLIP009403.jpg 760 | ..\Traing_Data\images\CLIP009398.jpg 761 | ..\Traing_Data\images\CLIP074F00616.jpg 762 | ..\Traing_Data\images\CLIP002991.jpg 763 | ..\Traing_Data\images\CLIP000663.jpg 764 | ..\Traing_Data\images\CLIP008908.jpg 765 | ..\Traing_Data\images\CLIP073F01050.jpg 766 | ..\Traing_Data\images\CLIP076F00534.jpg 767 | ..\Traing_Data\images\2115.jpg 768 | ..\Traing_Data\images\CLIP007605.jpg 769 | ..\Traing_Data\images\CLIP072F00315.jpg 770 | ..\Traing_Data\images\CLIP064F00186.jpg 771 | ..\Traing_Data\images\CLIP074F00458.jpg 772 | ..\Traing_Data\images\CLIP002003.jpg 773 | ..\Traing_Data\images\CLIP013338.jpg 774 | ..\Traing_Data\images\CLIP003822.jpg 775 | ..\Traing_Data\images\83.jpg 776 | ..\Traing_Data\images\CLIP075F01207.jpg 777 | ..\Traing_Data\images\CLIP003826.jpg 778 | ..\Traing_Data\images\CLIP077F00334.jpg 779 | ..\Traing_Data\images\CLIP069F00026.jpg 780 | ..\Traing_Data\images\CLIP008764.jpg 781 | ..\Traing_Data\images\CLIP075F00497.jpg 782 | ..\Traing_Data\images\CLIP071F00099.jpg 783 | ..\Traing_Data\images\2069.jpg 784 | ..\Traing_Data\images\CLIP006807.jpg 785 | ..\Traing_Data\images\CLIP074F00605.jpg 786 | ..\Traing_Data\images\3653.jpg 787 | ..\Traing_Data\images\CLIP071F00104.jpg 788 | ..\Traing_Data\images\1060.jpg 789 | ..\Traing_Data\images\CLIP075F00760.jpg 790 | ..\Traing_Data\images\CLIP009259.jpg 791 | ..\Traing_Data\images\CLIP073F00592.jpg 792 | ..\Traing_Data\images\CLIP075F00597.jpg 793 | ..\Traing_Data\images\CLIP074F00619.jpg 794 | ..\Traing_Data\images\CLIP013809.jpg 795 | ..\Traing_Data\images\1953.jpg 796 | ..\Traing_Data\images\CLIP010408.jpg 797 | ..\Traing_Data\images\CLIP072F00057.jpg 798 | ..\Traing_Data\images\CLIP009596.jpg 799 | ..\Traing_Data\images\CLIP006041.jpg 800 | ..\Traing_Data\images\CLIP074F00486.jpg 801 | ..\Traing_Data\images\CLIP073F00286.jpg 802 | ..\Traing_Data\images\CLIP066F00043.jpg 803 | ..\Traing_Data\images\CLIP007094.jpg 804 | ..\Traing_Data\images\CLIP074F00540.jpg 805 | ..\Traing_Data\images\CLIP005471.jpg 806 | ..\Traing_Data\images\3823.jpg 807 | ..\Traing_Data\images\CLIP012123.jpg 808 | ..\Traing_Data\images\3930.jpg 809 | ..\Traing_Data\images\CLIP079F00422.jpg 810 | ..\Traing_Data\images\CLIP007668.jpg 811 | ..\Traing_Data\images\3699.jpg 812 | ..\Traing_Data\images\CLIP078F03127.jpg 813 | ..\Traing_Data\images\CLIP003364.jpg 814 | ..\Traing_Data\images\CLIP069F00031.jpg 815 | ..\Traing_Data\images\2499.jpg 816 | ..\Traing_Data\images\3244.jpg 817 | ..\Traing_Data\images\CLIP075F00322.jpg 818 | ..\Traing_Data\images\CLIP007083.jpg 819 | ..\Traing_Data\images\CLIP012878.jpg 820 | ..\Traing_Data\images\CLIP001345.jpg 821 | ..\Traing_Data\images\CLIP072F00865.jpg 822 | ..\Traing_Data\images\CLIP004135.jpg 823 | ..\Traing_Data\images\CLIP065F00018.jpg 824 | ..\Traing_Data\images\CLIP073F00888.jpg 825 | ..\Traing_Data\images\3191.jpg 826 | ..\Traing_Data\images\CLIP071F00210.jpg 827 | ..\Traing_Data\images\CLIP069F00132.jpg 828 | ..\Traing_Data\images\CLIP079F00692.jpg 829 | ..\Traing_Data\images\3892.jpg 830 | ..\Traing_Data\images\CLIP007304.jpg 831 | ..\Traing_Data\images\CLIP064F00008.jpg 832 | ..\Traing_Data\images\CLIP078F03126.jpg 833 | ..\Traing_Data\images\CLIP068F00022.jpg 834 | ..\Traing_Data\images\CLIP074F00498.jpg 835 | ..\Traing_Data\images\3492.jpg 836 | ..\Traing_Data\images\CLIP077F00809.jpg 837 | ..\Traing_Data\images\CLIP000463.jpg 838 | ..\Traing_Data\images\CLIP074F00615.jpg 839 | ..\Traing_Data\images\CLIP072F00534.jpg 840 | ..\Traing_Data\images\2487.jpg 841 | ..\Traing_Data\images\CLIP006490.jpg 842 | ..\Traing_Data\images\CLIP007199.jpg 843 | ..\Traing_Data\images\CLIP073F00042.jpg 844 | ..\Traing_Data\images\CLIP064F00107.jpg 845 | ..\Traing_Data\images\CLIP011442.jpg 846 | ..\Traing_Data\images\3674.jpg 847 | ..\Traing_Data\images\CLIP071F00132.jpg 848 | ..\Traing_Data\images\CLIP000008.jpg 849 | ..\Traing_Data\images\CLIP002905.jpg 850 | ..\Traing_Data\images\CLIP072F00101.jpg 851 | ..\Traing_Data\images\CLIP009314.jpg 852 | ..\Traing_Data\images\2595.jpg 853 | ..\Traing_Data\images\CLIP004303.jpg 854 | ..\Traing_Data\images\CLIP013003.jpg 855 | ..\Traing_Data\images\CLIP072F00330.jpg 856 | ..\Traing_Data\images\CLIP064F00352.jpg 857 | ..\Traing_Data\images\93.jpg 858 | ..\Traing_Data\images\CLIP074F00407.jpg 859 | ..\Traing_Data\images\CLIP072F00197.jpg 860 | ..\Traing_Data\images\CLIP010570.jpg 861 | ..\Traing_Data\images\3816.jpg 862 | ..\Traing_Data\images\CLIP071F00113.jpg 863 | ..\Traing_Data\images\885.jpg 864 | ..\Traing_Data\images\CLIP008062.jpg 865 | ..\Traing_Data\images\CLIP073F00400.jpg 866 | ..\Traing_Data\images\CLIP002614.jpg 867 | ..\Traing_Data\images\CLIP074F00367.jpg 868 | ..\Traing_Data\images\CLIP009509.jpg 869 | ..\Traing_Data\images\3569.jpg 870 | ..\Traing_Data\images\CLIP007656.jpg 871 | ..\Traing_Data\images\CLIP077F00490.jpg 872 | ..\Traing_Data\images\CLIP006985.jpg 873 | ..\Traing_Data\images\CLIP012227.jpg 874 | ..\Traing_Data\images\CLIP000452.jpg 875 | ..\Traing_Data\images\3398.jpg 876 | ..\Traing_Data\images\CLIP073F00183.jpg 877 | ..\Traing_Data\images\CLIP068F00106.jpg 878 | ..\Traing_Data\images\CLIP064F00066.jpg 879 | ..\Traing_Data\images\3236.jpg 880 | ..\Traing_Data\images\864.jpg 881 | ..\Traing_Data\images\CLIP066F00056.jpg 882 | ..\Traing_Data\images\CLIP067F00006.jpg 883 | ..\Traing_Data\images\CLIP003469.jpg 884 | ..\Traing_Data\images\CLIP069F00162.jpg 885 | ..\Traing_Data\images\CLIP072F00107.jpg 886 | ..\Traing_Data\images\2969.jpg 887 | ..\Traing_Data\images\CLIP075F00522.jpg 888 | ..\Traing_Data\images\CLIP005189.jpg 889 | ..\Traing_Data\images\CLIP078F00502.jpg 890 | ..\Traing_Data\images\CLIP075F00475.jpg 891 | ..\Traing_Data\images\CLIP069F00066.jpg 892 | ..\Traing_Data\images\1558.jpg 893 | ..\Traing_Data\images\CLIP066F00053.jpg 894 | ..\Traing_Data\images\CLIP004722.jpg 895 | ..\Traing_Data\images\CLIP079F00520.jpg 896 | ..\Traing_Data\images\CLIP005442.jpg 897 | ..\Traing_Data\images\CLIP005843.jpg 898 | ..\Traing_Data\images\CLIP074F00457.jpg 899 | ..\Traing_Data\images\CLIP069F00069.jpg 900 | ..\Traing_Data\images\CLIP076F00243.jpg 901 | ..\Traing_Data\images\2215.jpg 902 | ..\Traing_Data\images\CLIP071F00106.jpg 903 | ..\Traing_Data\images\2372.jpg 904 | ..\Traing_Data\images\CLIP069F00169.jpg 905 | ..\Traing_Data\images\CLIP068F00075.jpg 906 | ..\Traing_Data\images\CLIP003963.jpg 907 | ..\Traing_Data\images\CLIP068F00041.jpg 908 | ..\Traing_Data\images\CLIP007986.jpg 909 | ..\Traing_Data\images\CLIP008548.jpg 910 | ..\Traing_Data\images\CLIP005696.jpg 911 | ..\Traing_Data\images\4128.jpg 912 | ..\Traing_Data\images\CLIP075F01215.jpg 913 | ..\Traing_Data\images\CLIP072F00758.jpg 914 | ..\Traing_Data\images\CLIP011261.jpg 915 | ..\Traing_Data\images\CLIP010474.jpg 916 | ..\Traing_Data\images\CLIP070F00073.jpg 917 | ..\Traing_Data\images\1059.jpg 918 | ..\Traing_Data\images\3673.jpg 919 | ..\Traing_Data\images\CLIP074F00292.jpg 920 | ..\Traing_Data\images\CLIP004980.jpg 921 | ..\Traing_Data\images\2389.jpg 922 | ..\Traing_Data\images\CLIP003463.jpg 923 | ..\Traing_Data\images\CLIP012148.jpg 924 | ..\Traing_Data\images\CLIP002495.jpg 925 | ..\Traing_Data\images\CLIP004118.jpg 926 | ..\Traing_Data\images\CLIP006268.jpg 927 | ..\Traing_Data\images\CLIP074F00352.jpg 928 | ..\Traing_Data\images\CLIP009147.jpg 929 | ..\Traing_Data\images\CLIP072F00777.jpg 930 | ..\Traing_Data\images\CLIP012566.jpg 931 | ..\Traing_Data\images\CLIP001630.jpg 932 | ..\Traing_Data\images\CLIP003331.jpg 933 | ..\Traing_Data\images\CLIP013004.jpg 934 | ..\Traing_Data\images\CLIP010859.jpg 935 | ..\Traing_Data\images\CLIP072F00591.jpg 936 | ..\Traing_Data\images\CLIP074F01072.jpg 937 | ..\Traing_Data\images\CLIP073F00613.jpg 938 | ..\Traing_Data\images\CLIP073F00683.jpg 939 | ..\Traing_Data\images\CLIP068F00001.jpg 940 | ..\Traing_Data\images\CLIP000973.jpg 941 | ..\Traing_Data\images\CLIP008775.jpg 942 | ..\Traing_Data\images\233.jpg 943 | ..\Traing_Data\images\2248.jpg 944 | ..\Traing_Data\images\CLIP073F00875.jpg 945 | ..\Traing_Data\images\CLIP003058.jpg 946 | ..\Traing_Data\images\CLIP071F00316.jpg 947 | ..\Traing_Data\images\CLIP075F00754.jpg 948 | ..\Traing_Data\images\CLIP003165.jpg 949 | ..\Traing_Data\images\CLIP071F00161.jpg 950 | ..\Traing_Data\images\CLIP004708.jpg 951 | ..\Traing_Data\images\CLIP077F00501.jpg 952 | ..\Traing_Data\images\CLIP012866.jpg 953 | ..\Traing_Data\images\2485.jpg 954 | ..\Traing_Data\images\CLIP072F00779.jpg 955 | ..\Traing_Data\images\CLIP077F00340.jpg 956 | ..\Traing_Data\images\CLIP074F00551.jpg 957 | ..\Traing_Data\images\CLIP075F00054.jpg 958 | ..\Traing_Data\images\3581.jpg 959 | ..\Traing_Data\images\CLIP072F00041.jpg 960 | ..\Traing_Data\images\CLIP072F00520.jpg 961 | ..\Traing_Data\images\CLIP014020.jpg 962 | ..\Traing_Data\images\CLIP066F00064.jpg 963 | ..\Traing_Data\images\CLIP075F00046.jpg 964 | ..\Traing_Data\images\1916.jpg 965 | ..\Traing_Data\images\CLIP072F00255.jpg 966 | ..\Traing_Data\images\CLIP072F00541.jpg 967 | ..\Traing_Data\images\CLIP005142.jpg 968 | ..\Traing_Data\images\CLIP010387.jpg 969 | ..\Traing_Data\images\1958.jpg 970 | ..\Traing_Data\images\CLIP075F00460.jpg 971 | ..\Traing_Data\images\CLIP076F00146.jpg 972 | ..\Traing_Data\images\CLIP068F00006.jpg 973 | ..\Traing_Data\images\CLIP009256.jpg 974 | ..\Traing_Data\images\CLIP002443.jpg 975 | ..\Traing_Data\images\CLIP003103.jpg 976 | ..\Traing_Data\images\CLIP073F00899.jpg 977 | ..\Traing_Data\images\CLIP075F00752.jpg 978 | ..\Traing_Data\images\CLIP076F00008.jpg 979 | ..\Traing_Data\images\CLIP077F00496.jpg 980 | ..\Traing_Data\images\CLIP007357.jpg 981 | ..\Traing_Data\images\3534.jpg 982 | ..\Traing_Data\images\3627.jpg 983 | ..\Traing_Data\images\CLIP075F00058.jpg 984 | ..\Traing_Data\images\CLIP013498.jpg 985 | ..\Traing_Data\images\CLIP068F00055.jpg 986 | ..\Traing_Data\images\CLIP011918.jpg 987 | ..\Traing_Data\images\CLIP077F00337.jpg 988 | ..\Traing_Data\images\2044.jpg 989 | ..\Traing_Data\images\CLIP064F00120.jpg 990 | ..\Traing_Data\images\CLIP010090.jpg 991 | ..\Traing_Data\images\CLIP080F00493.jpg 992 | ..\Traing_Data\images\3900.jpg 993 | ..\Traing_Data\images\CLIP070F00045.jpg 994 | ..\Traing_Data\images\CLIP069F00138.jpg 995 | ..\Traing_Data\images\CLIP073F00415.jpg 996 | ..\Traing_Data\images\CLIP077F00814.jpg 997 | ..\Traing_Data\images\CLIP002712.jpg 998 | ..\Traing_Data\images\CLIP000943.jpg 999 | ..\Traing_Data\images\2860.jpg 1000 | ..\Traing_Data\images\CLIP073F00028.jpg 1001 | ..\Traing_Data\images\CLIP073F00429.jpg 1002 | ..\Traing_Data\images\CLIP075F00671.jpg 1003 | ..\Traing_Data\images\CLIP003859.jpg 1004 | ..\Traing_Data\images\CLIP069F00049.jpg 1005 | ..\Traing_Data\images\CLIP072F00854.jpg 1006 | ..\Traing_Data\images\3364.jpg 1007 | ..\Traing_Data\images\260.jpg 1008 | ..\Traing_Data\images\CLIP010382.jpg 1009 | ..\Traing_Data\images\CLIP072F00551.jpg 1010 | ..\Traing_Data\images\CLIP076F00321.jpg 1011 | ..\Traing_Data\images\CLIP003595.jpg 1012 | ..\Traing_Data\images\CLIP011351.jpg 1013 | ..\Traing_Data\images\CLIP079F00459.jpg 1014 | ..\Traing_Data\images\CLIP003789.jpg 1015 | ..\Traing_Data\images\CLIP010290.jpg 1016 | ..\Traing_Data\images\CLIP069F00143.jpg 1017 | ..\Traing_Data\images\CLIP010897.jpg 1018 | ..\Traing_Data\images\CLIP072F00562.jpg 1019 | ..\Traing_Data\images\CLIP005584.jpg 1020 | ..\Traing_Data\images\CLIP069F00139.jpg 1021 | ..\Traing_Data\images\CLIP071F00134.jpg 1022 | ..\Traing_Data\images\CLIP072F00132.jpg 1023 | ..\Traing_Data\images\CLIP073F00603.jpg 1024 | ..\Traing_Data\images\CLIP072F00153.jpg 1025 | ..\Traing_Data\images\CLIP006778.jpg 1026 | ..\Traing_Data\images\CLIP008158.jpg 1027 | ..\Traing_Data\images\CLIP007980.jpg 1028 | ..\Traing_Data\images\CLIP075F00454.jpg 1029 | ..\Traing_Data\images\CLIP008128.jpg 1030 | ..\Traing_Data\images\CLIP072F00304.jpg 1031 | ..\Traing_Data\images\CLIP077F00307.jpg 1032 | ..\Traing_Data\images\2041.jpg 1033 | ..\Traing_Data\images\CLIP064F00386.jpg 1034 | ..\Traing_Data\images\CLIP074F00257.jpg 1035 | ..\Traing_Data\images\CLIP007196.jpg 1036 | ..\Traing_Data\images\CLIP074F00348.jpg 1037 | ..\Traing_Data\images\CLIP012831.jpg 1038 | ..\Traing_Data\images\CLIP072F00229.jpg 1039 | ..\Traing_Data\images\CLIP076F00005.jpg 1040 | ..\Traing_Data\images\CLIP013528.jpg 1041 | ..\Traing_Data\images\CLIP010849.jpg 1042 | ..\Traing_Data\images\CLIP078F00980.jpg 1043 | ..\Traing_Data\images\CLIP011534.jpg 1044 | ..\Traing_Data\images\CLIP009489.jpg 1045 | ..\Traing_Data\images\CLIP000877.jpg 1046 | ..\Traing_Data\images\3907.jpg 1047 | ..\Traing_Data\images\CLIP004300.jpg 1048 | ..\Traing_Data\images\CLIP007353.jpg 1049 | ..\Traing_Data\images\CLIP072F00146.jpg 1050 | ..\Traing_Data\images\1410.jpg 1051 | ..\Traing_Data\images\CLIP008355.jpg 1052 | ..\Traing_Data\images\CLIP071F00127.jpg 1053 | ..\Traing_Data\images\CLIP072F00510.jpg 1054 | ..\Traing_Data\images\CLIP064F00206.jpg 1055 | ..\Traing_Data\images\CLIP004991.jpg 1056 | ..\Traing_Data\images\CLIP072F00516.jpg 1057 | ..\Traing_Data\images\CLIP072F00105.jpg 1058 | ..\Traing_Data\images\CLIP011571.jpg 1059 | ..\Traing_Data\images\CLIP004169.jpg 1060 | ..\Traing_Data\images\CLIP075F00700.jpg 1061 | ..\Traing_Data\images\CLIP073F00284.jpg 1062 | ..\Traing_Data\images\463.jpg 1063 | ..\Traing_Data\images\CLIP004514.jpg 1064 | ..\Traing_Data\images\CLIP075F00043.jpg 1065 | ..\Traing_Data\images\CLIP000339.jpg 1066 | ..\Traing_Data\images\CLIP013235.jpg 1067 | ..\Traing_Data\images\CLIP001753.jpg 1068 | ..\Traing_Data\images\CLIP010110.jpg 1069 | ..\Traing_Data\images\CLIP012193.jpg 1070 | ..\Traing_Data\images\CLIP007461.jpg 1071 | ..\Traing_Data\images\CLIP071F00090.jpg 1072 | ..\Traing_Data\images\CLIP075F00483.jpg 1073 | ..\Traing_Data\images\CLIP074F00262.jpg 1074 | ..\Traing_Data\images\CLIP066F00033.jpg 1075 | ..\Traing_Data\images\CLIP071F00192.jpg 1076 | ..\Traing_Data\images\CLIP071F00177.jpg 1077 | ..\Traing_Data\images\CLIP068F00074.jpg 1078 | ..\Traing_Data\images\CLIP073F00628.jpg 1079 | ..\Traing_Data\images\CLIP072F00539.jpg 1080 | ..\Traing_Data\images\CLIP008724.jpg 1081 | ..\Traing_Data\images\CLIP073F00588.jpg 1082 | ..\Traing_Data\images\CLIP005625.jpg 1083 | ..\Traing_Data\images\CLIP006478.jpg 1084 | ..\Traing_Data\images\CLIP013808.jpg 1085 | ..\Traing_Data\images\CLIP072F00862.jpg 1086 | ..\Traing_Data\images\CLIP071F00269.jpg 1087 | ..\Traing_Data\images\CLIP005178.jpg 1088 | ..\Traing_Data\images\901.jpg 1089 | ..\Traing_Data\images\CLIP074F00202.jpg 1090 | ..\Traing_Data\images\CLIP012942.jpg 1091 | ..\Traing_Data\images\CLIP013843.jpg 1092 | ..\Traing_Data\images\CLIP000531.jpg 1093 | ..\Traing_Data\images\CLIP080F00033.jpg 1094 | ..\Traing_Data\images\42.jpg 1095 | ..\Traing_Data\images\CLIP074F00628.jpg 1096 | ..\Traing_Data\images\CLIP002141.jpg 1097 | ..\Traing_Data\images\CLIP007990.jpg 1098 | ..\Traing_Data\images\CLIP075F00040.jpg 1099 | ..\Traing_Data\images\CLIP074F00298.jpg 1100 | ..\Traing_Data\images\CLIP080F00027.jpg 1101 | ..\Traing_Data\images\CLIP064F00357.jpg 1102 | ..\Traing_Data\images\CLIP002745.jpg 1103 | ..\Traing_Data\images\CLIP074F00547.jpg 1104 | ..\Traing_Data\images\CLIP072F00766.jpg 1105 | ..\Traing_Data\images\CLIP005228.jpg 1106 | ..\Traing_Data\images\CLIP005468.jpg 1107 | ..\Traing_Data\images\CLIP073F00060.jpg 1108 | ..\Traing_Data\images\CLIP004179.jpg 1109 | ..\Traing_Data\images\CLIP078F00497.jpg 1110 | ..\Traing_Data\images\CLIP004838.jpg 1111 | ..\Traing_Data\images\CLIP064F00371.jpg 1112 | ..\Traing_Data\images\CLIP066F00022.jpg 1113 | ..\Traing_Data\images\CLIP074F00553.jpg 1114 | ..\Traing_Data\images\CLIP072F00768.jpg 1115 | ..\Traing_Data\images\CLIP071F00119.jpg 1116 | ..\Traing_Data\images\CLIP064F00024.jpg 1117 | ..\Traing_Data\images\CLIP073F00418.jpg 1118 | ..\Traing_Data\images\CLIP009784.jpg 1119 | ..\Traing_Data\images\1118.jpg 1120 | ..\Traing_Data\images\CLIP012230.jpg 1121 | ..\Traing_Data\images\CLIP005080.jpg 1122 | ..\Traing_Data\images\CLIP002305.jpg 1123 | ..\Traing_Data\images\3064.jpg 1124 | ..\Traing_Data\images\CLIP075F01627.jpg 1125 | ..\Traing_Data\images\CLIP073F00026.jpg 1126 | ..\Traing_Data\images\3223.jpg 1127 | ..\Traing_Data\images\1906.jpg 1128 | ..\Traing_Data\images\CLIP008596.jpg 1129 | ..\Traing_Data\images\CLIP075F02103.jpg 1130 | ..\Traing_Data\images\CLIP010826.jpg 1131 | ..\Traing_Data\images\CLIP012181.jpg 1132 | ..\Traing_Data\images\CLIP080F00028.jpg 1133 | ..\Traing_Data\images\CLIP069F00071.jpg 1134 | ..\Traing_Data\images\CLIP012357.jpg 1135 | ..\Traing_Data\images\CLIP010835.jpg 1136 | ..\Traing_Data\images\CLIP001272.jpg 1137 | ..\Traing_Data\images\CLIP003654.jpg 1138 | ..\Traing_Data\images\CLIP005044.jpg 1139 | ..\Traing_Data\images\CLIP075F00324.jpg 1140 | ..\Traing_Data\images\CLIP077F00002.jpg 1141 | ..\Traing_Data\images\CLIP067F00000.jpg 1142 | ..\Traing_Data\images\2377.jpg 1143 | ..\Traing_Data\images\CLIP002098.jpg 1144 | ..\Traing_Data\images\CLIP075F00510.jpg 1145 | ..\Traing_Data\images\CLIP070F00034.jpg 1146 | ..\Traing_Data\images\CLIP012724.jpg 1147 | ..\Traing_Data\images\CLIP064F00153.jpg 1148 | ..\Traing_Data\images\CLIP003208.jpg 1149 | ..\Traing_Data\images\CLIP073F01052.jpg 1150 | ..\Traing_Data\images\CLIP073F00892.jpg 1151 | ..\Traing_Data\images\CLIP073F00426.jpg 1152 | ..\Traing_Data\images\CLIP010883.jpg 1153 | ..\Traing_Data\images\CLIP068F00043.jpg 1154 | ..\Traing_Data\images\CLIP013658.jpg 1155 | ..\Traing_Data\images\CLIP006897.jpg 1156 | ..\Traing_Data\images\CLIP073F01035.jpg 1157 | ..\Traing_Data\images\CLIP013326.jpg 1158 | ..\Traing_Data\images\CLIP072F00108.jpg 1159 | ..\Traing_Data\images\CLIP074F00258.jpg 1160 | ..\Traing_Data\images\CLIP073F00187.jpg 1161 | ..\Traing_Data\images\4002.jpg 1162 | ..\Traing_Data\images\CLIP014072.jpg 1163 | ..\Traing_Data\images\CLIP073F01030.jpg 1164 | ..\Traing_Data\images\CLIP064F00098.jpg 1165 | ..\Traing_Data\images\1235.jpg 1166 | ..\Traing_Data\images\CLIP066F00133.jpg 1167 | ..\Traing_Data\images\CLIP072F00953.jpg 1168 | ..\Traing_Data\images\CLIP070F00054.jpg 1169 | ..\Traing_Data\images\CLIP073F00650.jpg 1170 | ..\Traing_Data\images\CLIP064F00116.jpg 1171 | ..\Traing_Data\images\CLIP004155.jpg 1172 | ..\Traing_Data\images\CLIP066F00031.jpg 1173 | ..\Traing_Data\images\CLIP002245.jpg 1174 | ..\Traing_Data\images\CLIP073F00905.jpg 1175 | ..\Traing_Data\images\CLIP073F00173.jpg 1176 | ..\Traing_Data\images\CLIP003750.jpg 1177 | ..\Traing_Data\images\CLIP067F00034.jpg 1178 | ..\Traing_Data\images\2721.jpg 1179 | ..\Traing_Data\images\CLIP001150.jpg 1180 | ..\Traing_Data\images\1082.jpg 1181 | ..\Traing_Data\images\CLIP072F00165.jpg 1182 | ..\Traing_Data\images\CLIP073F00674.jpg 1183 | ..\Traing_Data\images\CLIP006582.jpg 1184 | ..\Traing_Data\images\CLIP064F00393.jpg 1185 | ..\Traing_Data\images\CLIP069F00045.jpg 1186 | ..\Traing_Data\images\1284.jpg 1187 | ..\Traing_Data\images\CLIP074F00618.jpg 1188 | ..\Traing_Data\images\CLIP074F00631.jpg 1189 | ..\Traing_Data\images\CLIP007415.jpg 1190 | ..\Traing_Data\images\CLIP073F00000.jpg 1191 | ..\Traing_Data\images\1409.jpg 1192 | ..\Traing_Data\images\CLIP001065.jpg 1193 | ..\Traing_Data\images\CLIP072F00034.jpg 1194 | ..\Traing_Data\images\CLIP005977.jpg 1195 | ..\Traing_Data\images\2112.jpg 1196 | ..\Traing_Data\images\CLIP074F00622.jpg 1197 | ..\Traing_Data\images\CLIP071F00262.jpg 1198 | ..\Traing_Data\images\CLIP064F00029.jpg 1199 | ..\Traing_Data\images\CLIP007098.jpg 1200 | ..\Traing_Data\images\CLIP071F00289.jpg 1201 | ..\Traing_Data\images\CLIP075F00601.jpg 1202 | ..\Traing_Data\images\CLIP013117.jpg 1203 | ..\Traing_Data\images\CLIP069F00055.jpg 1204 | ..\Traing_Data\images\CLIP076F00037.jpg 1205 | ..\Traing_Data\images\2439.jpg 1206 | ..\Traing_Data\images\CLIP006243.jpg 1207 | ..\Traing_Data\images\CLIP073F00420.jpg 1208 | ..\Traing_Data\images\CLIP072F00299.jpg 1209 | ..\Traing_Data\images\4066.jpg 1210 | ..\Traing_Data\images\CLIP070F00064.jpg 1211 | ..\Traing_Data\images\CLIP009086.jpg 1212 | ..\Traing_Data\images\CLIP003052.jpg 1213 | ..\Traing_Data\images\CLIP077F01410.jpg 1214 | ..\Traing_Data\images\CLIP000172.jpg 1215 | ..\Traing_Data\images\CLIP005155.jpg 1216 | ..\Traing_Data\images\CLIP077F00489.jpg 1217 | ..\Traing_Data\images\CLIP079F00457.jpg 1218 | ..\Traing_Data\images\CLIP005631.jpg 1219 | ..\Traing_Data\images\3088.jpg 1220 | ..\Traing_Data\images\CLIP074F00284.jpg 1221 | ..\Traing_Data\images\CLIP073F00586.jpg 1222 | ..\Traing_Data\images\CLIP073F01315.jpg 1223 | ..\Traing_Data\images\CLIP064F00189.jpg 1224 | ..\Traing_Data\images\CLIP004932.jpg 1225 | ..\Traing_Data\images\CLIP074F00203.jpg 1226 | ..\Traing_Data\images\CLIP008401.jpg 1227 | ..\Traing_Data\images\CLIP072F00773.jpg 1228 | ..\Traing_Data\images\CLIP000224.jpg 1229 | ..\Traing_Data\images\CLIP069F00057.jpg 1230 | ..\Traing_Data\images\CLIP072F00186.jpg 1231 | ..\Traing_Data\images\CLIP078F00701.jpg 1232 | ..\Traing_Data\images\3859.jpg 1233 | ..\Traing_Data\images\CLIP008474.jpg 1234 | ..\Traing_Data\images\CLIP001278.jpg 1235 | ..\Traing_Data\images\CLIP011833.jpg 1236 | ..\Traing_Data\images\CLIP010937.jpg 1237 | ..\Traing_Data\images\CLIP007613.jpg 1238 | ..\Traing_Data\images\CLIP002728.jpg 1239 | ..\Traing_Data\images\CLIP002867.jpg 1240 | ..\Traing_Data\images\CLIP069F00180.jpg 1241 | ..\Traing_Data\images\433.jpg 1242 | ..\Traing_Data\images\3510.jpg 1243 | ..\Traing_Data\images\3839.jpg 1244 | ..\Traing_Data\images\CLIP006623.jpg 1245 | ..\Traing_Data\images\4145.jpg 1246 | ..\Traing_Data\images\CLIP008330.jpg 1247 | ..\Traing_Data\images\CLIP008058.jpg 1248 | ..\Traing_Data\images\CLIP012615.jpg 1249 | ..\Traing_Data\images\CLIP073F00906.jpg 1250 | ..\Traing_Data\images\CLIP080F00000.jpg 1251 | ..\Traing_Data\images\CLIP064F00225.jpg 1252 | ..\Traing_Data\images\3903.jpg 1253 | ..\Traing_Data\images\CLIP001472.jpg 1254 | ..\Traing_Data\images\CLIP073F01021.jpg 1255 | ..\Traing_Data\images\3978.jpg 1256 | ..\Traing_Data\images\CLIP071F00016.jpg 1257 | ..\Traing_Data\images\CLIP012106.jpg 1258 | ..\Traing_Data\images\CLIP071F00179.jpg 1259 | ..\Traing_Data\images\CLIP011348.jpg 1260 | ..\Traing_Data\images\650.jpg 1261 | ..\Traing_Data\images\CLIP072F00778.jpg 1262 | ..\Traing_Data\images\CLIP012347.jpg 1263 | ..\Traing_Data\images\CLIP002900.jpg 1264 | ..\Traing_Data\images\CLIP064F00236.jpg 1265 | ..\Traing_Data\images\CLIP004267.jpg 1266 | ..\Traing_Data\images\CLIP072F00332.jpg 1267 | ..\Traing_Data\images\CLIP072F00505.jpg 1268 | ..\Traing_Data\images\CLIP071F00125.jpg 1269 | ..\Traing_Data\images\CLIP073F00664.jpg 1270 | ..\Traing_Data\images\CLIP073F00061.jpg 1271 | ..\Traing_Data\images\CLIP080F00030.jpg 1272 | ..\Traing_Data\images\CLIP068F00018.jpg 1273 | ..\Traing_Data\images\CLIP007283.jpg 1274 | ..\Traing_Data\images\CLIP072F00072.jpg 1275 | ..\Traing_Data\images\CLIP011809.jpg 1276 | ..\Traing_Data\images\CLIP006635.jpg 1277 | ..\Traing_Data\images\3522.jpg 1278 | ..\Traing_Data\images\CLIP003955.jpg 1279 | ..\Traing_Data\images\CLIP000237.jpg 1280 | ..\Traing_Data\images\CLIP073F01025.jpg 1281 | ..\Traing_Data\images\CLIP002711.jpg 1282 | ..\Traing_Data\images\1291.jpg 1283 | ..\Traing_Data\images\CLIP079F00549.jpg 1284 | ..\Traing_Data\images\CLIP072F00044.jpg 1285 | ..\Traing_Data\images\4019.jpg 1286 | ..\Traing_Data\images\CLIP013381.jpg 1287 | ..\Traing_Data\images\CLIP004539.jpg 1288 | ..\Traing_Data\images\CLIP071F00102.jpg 1289 | ..\Traing_Data\images\CLIP073F00789.jpg 1290 | ..\Traing_Data\images\CLIP071F00271.jpg 1291 | ..\Traing_Data\images\929.jpg 1292 | ..\Traing_Data\images\CLIP072F00305.jpg 1293 | ..\Traing_Data\images\CLIP013487.jpg 1294 | ..\Traing_Data\images\CLIP071F00254.jpg 1295 | ..\Traing_Data\images\CLIP013146.jpg 1296 | ..\Traing_Data\images\CLIP067F00003.jpg 1297 | ..\Traing_Data\images\CLIP009420.jpg 1298 | ..\Traing_Data\images\CLIP064F00391.jpg 1299 | ..\Traing_Data\images\CLIP064F00376.jpg 1300 | ..\Traing_Data\images\CLIP001578.jpg 1301 | ..\Traing_Data\images\CLIP066F00131.jpg 1302 | ..\Traing_Data\images\CLIP003116.jpg 1303 | ..\Traing_Data\images\CLIP074F00287.jpg 1304 | ..\Traing_Data\images\CLIP071F00244.jpg 1305 | ..\Traing_Data\images\CLIP064F00401.jpg 1306 | ..\Traing_Data\images\CLIP001955.jpg 1307 | ..\Traing_Data\images\CLIP013935.jpg 1308 | ..\Traing_Data\images\2582.jpg 1309 | ..\Traing_Data\images\CLIP079F00548.jpg 1310 | ..\Traing_Data\images\CLIP075F00512.jpg 1311 | ..\Traing_Data\images\CLIP005478.jpg 1312 | ..\Traing_Data\images\CLIP072F00518.jpg 1313 | ..\Traing_Data\images\CLIP078F00977.jpg 1314 | ..\Traing_Data\images\3202.jpg 1315 | ..\Traing_Data\images\CLIP009444.jpg 1316 | ..\Traing_Data\images\CLIP012365.jpg 1317 | ..\Traing_Data\images\CLIP010312.jpg 1318 | ..\Traing_Data\images\3205.jpg 1319 | ..\Traing_Data\images\CLIP073F00798.jpg 1320 | ..\Traing_Data\images\CLIP010827.jpg 1321 | ..\Traing_Data\images\CLIP010298.jpg 1322 | ..\Traing_Data\images\CLIP064F00311.jpg 1323 | ..\Traing_Data\images\CLIP075F01377.jpg 1324 | ..\Traing_Data\images\CLIP007730.jpg 1325 | ..\Traing_Data\images\CLIP006670.jpg 1326 | ..\Traing_Data\images\CLIP004699.jpg 1327 | ..\Traing_Data\images\CLIP006456.jpg 1328 | ..\Traing_Data\images\CLIP011089.jpg 1329 | ..\Traing_Data\images\CLIP009555.jpg 1330 | ..\Traing_Data\images\CLIP001736.jpg 1331 | ..\Traing_Data\images\CLIP071F00183.jpg 1332 | ..\Traing_Data\images\CLIP001028.jpg 1333 | ..\Traing_Data\images\CLIP074F00000.jpg 1334 | ..\Traing_Data\images\CLIP003266.jpg 1335 | ..\Traing_Data\images\1980.jpg 1336 | ..\Traing_Data\images\CLIP008774.jpg 1337 | ..\Traing_Data\images\CLIP000212.jpg 1338 | ..\Traing_Data\images\CLIP073F00891.jpg 1339 | ..\Traing_Data\images\CLIP072F00144.jpg 1340 | ..\Traing_Data\images\CLIP001838.jpg 1341 | ..\Traing_Data\images\CLIP073F00647.jpg 1342 | ..\Traing_Data\images\CLIP069F00160.jpg 1343 | ..\Traing_Data\images\CLIP005055.jpg 1344 | ..\Traing_Data\images\CLIP003376.jpg 1345 | ..\Traing_Data\images\CLIP011291.jpg 1346 | ..\Traing_Data\images\CLIP072F00138.jpg 1347 | ..\Traing_Data\images\CLIP012576.jpg 1348 | ..\Traing_Data\images\CLIP004931.jpg 1349 | ..\Traing_Data\images\CLIP073F01012.jpg 1350 | ..\Traing_Data\images\CLIP012693.jpg 1351 | ..\Traing_Data\images\CLIP075F00426.jpg 1352 | ..\Traing_Data\images\CLIP001825.jpg 1353 | ..\Traing_Data\images\CLIP012217.jpg 1354 | ..\Traing_Data\images\CLIP003330.jpg 1355 | ..\Traing_Data\images\CLIP069F00006.jpg 1356 | ..\Traing_Data\images\CLIP013727.jpg 1357 | ..\Traing_Data\images\CLIP072F00333.jpg 1358 | ..\Traing_Data\images\CLIP068F00045.jpg 1359 | ..\Traing_Data\images\CLIP066F00135.jpg 1360 | ..\Traing_Data\images\1223.jpg 1361 | ..\Traing_Data\images\CLIP008096.jpg 1362 | ..\Traing_Data\images\CLIP072F00490.jpg 1363 | ..\Traing_Data\images\CLIP064F00135.jpg 1364 | ..\Traing_Data\images\CLIP072F00015.jpg 1365 | ..\Traing_Data\images\CLIP072F00580.jpg 1366 | ..\Traing_Data\images\CLIP072F00493.jpg 1367 | ..\Traing_Data\images\CLIP074F00276.jpg 1368 | ..\Traing_Data\images\CLIP007131.jpg 1369 | ..\Traing_Data\images\CLIP073F00212.jpg 1370 | ..\Traing_Data\images\CLIP064F00347.jpg 1371 | ..\Traing_Data\images\CLIP069F00188.jpg 1372 | ..\Traing_Data\images\CLIP011478.jpg 1373 | ..\Traing_Data\images\CLIP071F00182.jpg 1374 | ..\Traing_Data\images\1211.jpg 1375 | ..\Traing_Data\images\CLIP012100.jpg 1376 | ..\Traing_Data\images\CLIP072F00530.jpg 1377 | ..\Traing_Data\images\CLIP009041.jpg 1378 | ..\Traing_Data\images\CLIP004595.jpg 1379 | ..\Traing_Data\images\918.jpg 1380 | ..\Traing_Data\images\CLIP002441.jpg 1381 | ..\Traing_Data\images\CLIP009250.jpg 1382 | ..\Traing_Data\images\CLIP007760.jpg 1383 | ..\Traing_Data\images\CLIP071F00116.jpg 1384 | ..\Traing_Data\images\2412.jpg 1385 | ..\Traing_Data\images\CLIP012996.jpg 1386 | ..\Traing_Data\images\CLIP010188.jpg 1387 | ..\Traing_Data\images\CLIP075F00520.jpg 1388 | ..\Traing_Data\images\CLIP068F00047.jpg 1389 | ..\Traing_Data\images\CLIP069F00142.jpg 1390 | ..\Traing_Data\images\CLIP066F00038.jpg 1391 | ..\Traing_Data\images\CLIP068F00087.jpg 1392 | ..\Traing_Data\images\CLIP064F00378.jpg 1393 | ..\Traing_Data\images\CLIP007867.jpg 1394 | ..\Traing_Data\images\270.jpg 1395 | ..\Traing_Data\images\CLIP000420.jpg 1396 | ..\Traing_Data\images\CLIP010767.jpg 1397 | ..\Traing_Data\images\CLIP071F00126.jpg 1398 | ..\Traing_Data\images\CLIP072F00190.jpg 1399 | ..\Traing_Data\images\2384.jpg 1400 | ..\Traing_Data\images\CLIP010732.jpg 1401 | ..\Traing_Data\images\CLIP074F00528.jpg 1402 | ..\Traing_Data\images\3098.jpg 1403 | ..\Traing_Data\images\CLIP066F00020.jpg 1404 | ..\Traing_Data\images\CLIP010804.jpg 1405 | ..\Traing_Data\images\CLIP072F00929.jpg 1406 | ..\Traing_Data\images\3352.jpg 1407 | ..\Traing_Data\images\CLIP069F00082.jpg 1408 | ..\Traing_Data\images\CLIP003029.jpg 1409 | ..\Traing_Data\images\CLIP010138.jpg 1410 | ..\Traing_Data\images\CLIP068F00038.jpg 1411 | ..\Traing_Data\images\CLIP072F00084.jpg 1412 | ..\Traing_Data\images\CLIP072F00529.jpg 1413 | ..\Traing_Data\images\CLIP073F00399.jpg 1414 | ..\Traing_Data\images\CLIP075F00757.jpg 1415 | ..\Traing_Data\images\CLIP072F00572.jpg 1416 | ..\Traing_Data\images\CLIP073F00428.jpg 1417 | ..\Traing_Data\images\CLIP010615.jpg 1418 | ..\Traing_Data\images\CLIP072F00858.jpg 1419 | ..\Traing_Data\images\CLIP064F00053.jpg 1420 | ..\Traing_Data\images\CLIP072F00090.jpg 1421 | ..\Traing_Data\images\CLIP077F01868.jpg 1422 | ..\Traing_Data\images\3520.jpg 1423 | ..\Traing_Data\images\1338.jpg 1424 | ..\Traing_Data\images\CLIP072F00000.jpg 1425 | ..\Traing_Data\images\2612.jpg 1426 | ..\Traing_Data\images\CLIP069F00086.jpg 1427 | ..\Traing_Data\images\CLIP010009.jpg 1428 | ..\Traing_Data\images\CLIP072F00037.jpg 1429 | ..\Traing_Data\images\CLIP072F00549.jpg 1430 | ..\Traing_Data\images\CLIP076F00298.jpg 1431 | ..\Traing_Data\images\CLIP012329.jpg 1432 | ..\Traing_Data\images\CLIP069F00106.jpg 1433 | ..\Traing_Data\images\CLIP072F00555.jpg 1434 | ..\Traing_Data\images\CLIP002617.jpg 1435 | ..\Traing_Data\images\CLIP080F00041.jpg 1436 | ..\Traing_Data\images\CLIP072F00869.jpg 1437 | ..\Traing_Data\images\CLIP003352.jpg 1438 | ..\Traing_Data\images\CLIP008797.jpg 1439 | ..\Traing_Data\images\CLIP012059.jpg 1440 | ..\Traing_Data\images\CLIP074F01069.jpg 1441 | ..\Traing_Data\images\CLIP072F00954.jpg 1442 | ..\Traing_Data\images\CLIP005595.jpg 1443 | ..\Traing_Data\images\CLIP008767.jpg 1444 | ..\Traing_Data\images\CLIP074F00264.jpg 1445 | ..\Traing_Data\images\500.jpg 1446 | ..\Traing_Data\images\CLIP075F01376.jpg 1447 | ..\Traing_Data\images\CLIP010769.jpg 1448 | ..\Traing_Data\images\CLIP073F00663.jpg 1449 | ..\Traing_Data\images\3685.jpg 1450 | ..\Traing_Data\images\CLIP072F00602.jpg 1451 | ..\Traing_Data\images\CLIP077F01298.jpg 1452 | ..\Traing_Data\images\CLIP074F00728.jpg 1453 | ..\Traing_Data\images\CLIP072F00076.jpg 1454 | ..\Traing_Data\images\CLIP068F00091.jpg 1455 | ..\Traing_Data\images\CLIP064F00207.jpg 1456 | ..\Traing_Data\images\CLIP071F00079.jpg 1457 | ..\Traing_Data\images\CLIP005269.jpg 1458 | ..\Traing_Data\images\CLIP073F00213.jpg 1459 | ..\Traing_Data\images\CLIP008962.jpg 1460 | ..\Traing_Data\images\CLIP072F00121.jpg 1461 | ..\Traing_Data\images\CLIP076F00302.jpg 1462 | ..\Traing_Data\images\CLIP072F00191.jpg 1463 | ..\Traing_Data\images\CLIP069F00010.jpg 1464 | ..\Traing_Data\images\CLIP072F00502.jpg 1465 | ..\Traing_Data\images\CLIP075F00367.jpg 1466 | ..\Traing_Data\images\CLIP067F00035.jpg 1467 | ..\Traing_Data\images\CLIP000312.jpg 1468 | ..\Traing_Data\images\CLIP068F00049.jpg 1469 | ..\Traing_Data\images\CLIP074F00363.jpg 1470 | ..\Traing_Data\images\3509.jpg 1471 | ..\Traing_Data\images\2286.jpg 1472 | ..\Traing_Data\images\CLIP072F00232.jpg 1473 | ..\Traing_Data\images\CLIP069F00170.jpg 1474 | ..\Traing_Data\images\2468.jpg 1475 | ..\Traing_Data\images\454.jpg 1476 | ..\Traing_Data\images\CLIP074F00488.jpg 1477 | ..\Traing_Data\images\96.jpg 1478 | ..\Traing_Data\images\CLIP073F01316.jpg 1479 | ..\Traing_Data\images\CLIP009696.jpg 1480 | ..\Traing_Data\images\CLIP079F00517.jpg 1481 | ..\Traing_Data\images\CLIP076F00661.jpg 1482 | ..\Traing_Data\images\CLIP074F00420.jpg 1483 | ..\Traing_Data\images\CLIP071F00015.jpg 1484 | ..\Traing_Data\images\CLIP007413.jpg 1485 | ..\Traing_Data\images\CLIP013064.jpg 1486 | ..\Traing_Data\images\CLIP011882.jpg 1487 | ..\Traing_Data\images\CLIP013159.jpg 1488 | ..\Traing_Data\images\3939.jpg 1489 | ..\Traing_Data\images\CLIP064F00178.jpg 1490 | ..\Traing_Data\images\CLIP009171.jpg 1491 | ..\Traing_Data\images\CLIP011337.jpg 1492 | ..\Traing_Data\images\CLIP011424.jpg 1493 | ..\Traing_Data\images\CLIP009360.jpg 1494 | ..\Traing_Data\images\CLIP075F00464.jpg 1495 | ..\Traing_Data\images\2965.jpg 1496 | ..\Traing_Data\images\CLIP006633.jpg 1497 | ..\Traing_Data\images\CLIP006979.jpg 1498 | ..\Traing_Data\images\CLIP001868.jpg 1499 | ..\Traing_Data\images\CLIP008911.jpg 1500 | ..\Traing_Data\images\CLIP007660.jpg 1501 | ..\Traing_Data\images\CLIP003900.jpg 1502 | ..\Traing_Data\images\3087.jpg 1503 | ..\Traing_Data\images\CLIP076F00282.jpg 1504 | ..\Traing_Data\images\CLIP069F00032.jpg 1505 | ..\Traing_Data\images\CLIP071F00203.jpg 1506 | ..\Traing_Data\images\CLIP076F00297.jpg 1507 | ..\Traing_Data\images\CLIP001031.jpg 1508 | ..\Traing_Data\images\CLIP072F00942.jpg 1509 | ..\Traing_Data\images\582.jpg 1510 | ..\Traing_Data\images\CLIP003962.jpg 1511 | ..\Traing_Data\images\681.jpg 1512 | ..\Traing_Data\images\CLIP004203.jpg 1513 | ..\Traing_Data\images\CLIP011573.jpg 1514 | ..\Traing_Data\images\CLIP005559.jpg 1515 | ..\Traing_Data\images\CLIP073F00631.jpg 1516 | ..\Traing_Data\images\CLIP002758.jpg 1517 | ..\Traing_Data\images\CLIP004714.jpg 1518 | ..\Traing_Data\images\CLIP003777.jpg 1519 | ..\Traing_Data\images\2991.jpg 1520 | ..\Traing_Data\images\CLIP077F00474.jpg 1521 | ..\Traing_Data\images\CLIP064F00238.jpg 1522 | ..\Traing_Data\images\CLIP075F00498.jpg 1523 | ..\Traing_Data\images\CLIP066F00066.jpg 1524 | ..\Traing_Data\images\CLIP064F00028.jpg 1525 | ..\Traing_Data\images\1210.jpg 1526 | ..\Traing_Data\images\CLIP072F00765.jpg 1527 | ..\Traing_Data\images\CLIP073F00006.jpg 1528 | ..\Traing_Data\images\CLIP072F00119.jpg 1529 | ..\Traing_Data\images\CLIP075F00544.jpg 1530 | ..\Traing_Data\images\CLIP001005.jpg 1531 | ..\Traing_Data\images\CLIP074F00448.jpg 1532 | ..\Traing_Data\images\CLIP074F00539.jpg 1533 | ..\Traing_Data\images\CLIP010265.jpg 1534 | ..\Traing_Data\images\CLIP071F00242.jpg 1535 | ..\Traing_Data\images\CLIP077F01495.jpg 1536 | ..\Traing_Data\images\CLIP066F00019.jpg 1537 | ..\Traing_Data\images\CLIP004783.jpg 1538 | ..\Traing_Data\images\CLIP072F00054.jpg 1539 | ..\Traing_Data\images\CLIP012857.jpg 1540 | ..\Traing_Data\images\CLIP071F00021.jpg 1541 | ..\Traing_Data\images\CLIP079F00694.jpg 1542 | ..\Traing_Data\images\CLIP076F00535.jpg 1543 | ..\Traing_Data\images\CLIP073F00655.jpg 1544 | ..\Traing_Data\images\CLIP077F01405.jpg 1545 | ..\Traing_Data\images\CLIP009749.jpg 1546 | ..\Traing_Data\images\2234.jpg 1547 | ..\Traing_Data\images\CLIP008503.jpg 1548 | ..\Traing_Data\images\CLIP072F00609.jpg 1549 | ..\Traing_Data\images\CLIP074F00418.jpg 1550 | ..\Traing_Data\images\CLIP072F00957.jpg 1551 | ..\Traing_Data\images\CLIP005316.jpg 1552 | ..\Traing_Data\images\CLIP007172.jpg 1553 | ..\Traing_Data\images\CLIP009205.jpg 1554 | ..\Traing_Data\images\CLIP010797.jpg 1555 | ..\Traing_Data\images\CLIP076F00326.jpg 1556 | ..\Traing_Data\images\CLIP064F00016.jpg 1557 | ..\Traing_Data\images\CLIP001540.jpg 1558 | ..\Traing_Data\images\CLIP077F00512.jpg 1559 | ..\Traing_Data\images\CLIP012533.jpg 1560 | ..\Traing_Data\images\CLIP075F00458.jpg 1561 | ..\Traing_Data\images\CLIP075F00428.jpg 1562 | ..\Traing_Data\images\CLIP074F00451.jpg 1563 | ..\Traing_Data\images\3734.jpg 1564 | ..\Traing_Data\images\CLIP002196.jpg 1565 | ..\Traing_Data\images\CLIP006245.jpg 1566 | ..\Traing_Data\images\CLIP069F00122.jpg 1567 | ..\Traing_Data\images\CLIP071F00013.jpg 1568 | ..\Traing_Data\images\CLIP076F00280.jpg 1569 | ..\Traing_Data\images\CLIP073F01311.jpg 1570 | ..\Traing_Data\images\CLIP073F00419.jpg 1571 | ..\Traing_Data\images\CLIP008607.jpg 1572 | ..\Traing_Data\images\CLIP073F00665.jpg 1573 | ..\Traing_Data\images\CLIP068F00085.jpg 1574 | ..\Traing_Data\images\CLIP004487.jpg 1575 | ..\Traing_Data\images\CLIP064F00239.jpg 1576 | ..\Traing_Data\images\CLIP074F00358.jpg 1577 | ..\Traing_Data\images\CLIP076F00000.jpg 1578 | ..\Traing_Data\images\707.jpg 1579 | ..\Traing_Data\images\781.jpg 1580 | ..\Traing_Data\images\CLIP069F00016.jpg 1581 | ..\Traing_Data\images\CLIP012498.jpg 1582 | ..\Traing_Data\images\CLIP072F00060.jpg 1583 | ..\Traing_Data\images\CLIP074F00485.jpg 1584 | ..\Traing_Data\images\CLIP069F00111.jpg 1585 | ..\Traing_Data\images\CLIP006000.jpg 1586 | ..\Traing_Data\images\CLIP072F00183.jpg 1587 | ..\Traing_Data\images\2976.jpg 1588 | ..\Traing_Data\images\CLIP066F00068.jpg 1589 | ..\Traing_Data\images\CLIP077F00484.jpg 1590 | ..\Traing_Data\images\CLIP014012.jpg 1591 | ..\Traing_Data\images\CLIP068F00090.jpg 1592 | ..\Traing_Data\images\CLIP073F00627.jpg 1593 | ..\Traing_Data\images\2979.jpg 1594 | ..\Traing_Data\images\1729.jpg 1595 | ..\Traing_Data\images\CLIP072F00080.jpg 1596 | ..\Traing_Data\images\CLIP072F00533.jpg 1597 | ..\Traing_Data\images\CLIP080F00451.jpg 1598 | ..\Traing_Data\images\4084.jpg 1599 | ..\Traing_Data\images\CLIP073F00406.jpg 1600 | ..\Traing_Data\images\CLIP073F01013.jpg 1601 | ..\Traing_Data\images\CLIP013988.jpg 1602 | ..\Traing_Data\images\CLIP073F00621.jpg 1603 | ..\Traing_Data\images\CLIP075F01374.jpg 1604 | ..\Traing_Data\images\CLIP071F00156.jpg 1605 | ..\Traing_Data\images\CLIP068F00053.jpg 1606 | ..\Traing_Data\images\CLIP072F00542.jpg 1607 | ..\Traing_Data\images\CLIP011473.jpg 1608 | ..\Traing_Data\images\CLIP012258.jpg 1609 | ..\Traing_Data\images\CLIP065F00001.jpg 1610 | ..\Traing_Data\images\3555.jpg 1611 | ..\Traing_Data\images\CLIP067F00030.jpg 1612 | ..\Traing_Data\images\CLIP009935.jpg 1613 | ..\Traing_Data\images\CLIP068F00097.jpg 1614 | ..\Traing_Data\images\CLIP072F00055.jpg 1615 | ..\Traing_Data\images\CLIP068F00024.jpg 1616 | ..\Traing_Data\images\CLIP073F00289.jpg 1617 | ..\Traing_Data\images\CLIP001434.jpg 1618 | ..\Traing_Data\images\CLIP072F00521.jpg 1619 | ..\Traing_Data\images\CLIP007220.jpg 1620 | ..\Traing_Data\images\CLIP067F00001.jpg 1621 | ..\Traing_Data\images\CLIP073F01020.jpg 1622 | ..\Traing_Data\images\CLIP013918.jpg 1623 | ..\Traing_Data\images\CLIP073F00876.jpg 1624 | ..\Traing_Data\images\CLIP011292.jpg 1625 | ..\Traing_Data\images\CLIP077F00807.jpg 1626 | ..\Traing_Data\images\2367.jpg 1627 | ..\Traing_Data\images\CLIP072F00910.jpg 1628 | ..\Traing_Data\images\CLIP069F00146.jpg 1629 | ..\Traing_Data\images\CLIP013482.jpg 1630 | ..\Traing_Data\images\2296.jpg 1631 | ..\Traing_Data\images\CLIP001000.jpg 1632 | ..\Traing_Data\images\CLIP004563.jpg 1633 | ..\Traing_Data\images\CLIP005415.jpg 1634 | ..\Traing_Data\images\CLIP005182.jpg 1635 | ..\Traing_Data\images\761.jpg 1636 | ..\Traing_Data\images\CLIP072F00158.jpg 1637 | ..\Traing_Data\images\CLIP079F01733.jpg 1638 | ..\Traing_Data\images\CLIP001263.jpg 1639 | ..\Traing_Data\images\CLIP000646.jpg 1640 | ..\Traing_Data\images\CLIP072F00605.jpg 1641 | ..\Traing_Data\images\CLIP001783.jpg 1642 | ..\Traing_Data\images\CLIP068F00050.jpg 1643 | ..\Traing_Data\images\CLIP072F00867.jpg 1644 | ..\Traing_Data\images\CLIP074F00268.jpg 1645 | ..\Traing_Data\images\CLIP008304.jpg 1646 | ..\Traing_Data\images\CLIP007608.jpg 1647 | ..\Traing_Data\images\CLIP074F00356.jpg 1648 | ..\Traing_Data\images\CLIP014039.jpg 1649 | ..\Traing_Data\images\CLIP003729.jpg 1650 | ..\Traing_Data\images\CLIP072F00860.jpg 1651 | ..\Traing_Data\images\CLIP010349.jpg 1652 | ..\Traing_Data\images\513.jpg 1653 | ..\Traing_Data\images\CLIP075F00471.jpg 1654 | ..\Traing_Data\images\CLIP013538.jpg 1655 | ..\Traing_Data\images\CLIP071F00307.jpg 1656 | ..\Traing_Data\images\CLIP012664.jpg 1657 | ..\Traing_Data\images\CLIP012946.jpg 1658 | ..\Traing_Data\images\CLIP073F01213.jpg 1659 | ..\Traing_Data\images\CLIP008944.jpg 1660 | ..\Traing_Data\images\CLIP003891.jpg 1661 | ..\Traing_Data\images\CLIP071F00229.jpg 1662 | ..\Traing_Data\images\1403.jpg 1663 | ..\Traing_Data\images\2746.jpg 1664 | ..\Traing_Data\images\CLIP002028.jpg 1665 | ..\Traing_Data\images\CLIP014093.jpg 1666 | ..\Traing_Data\images\561.jpg 1667 | ..\Traing_Data\images\CLIP073F00615.jpg 1668 | ..\Traing_Data\images\CLIP064F00173.jpg 1669 | ..\Traing_Data\images\CLIP000442.jpg 1670 | ..\Traing_Data\images\CLIP071F00222.jpg 1671 | ..\Traing_Data\images\CLIP075F01208.jpg 1672 | ..\Traing_Data\images\CLIP072F00045.jpg 1673 | ..\Traing_Data\images\CLIP004614.jpg 1674 | ..\Traing_Data\images\CLIP001770.jpg 1675 | ..\Traing_Data\images\666.jpg 1676 | ..\Traing_Data\images\CLIP068F00100.jpg 1677 | ..\Traing_Data\images\CLIP007003.jpg 1678 | ..\Traing_Data\images\CLIP075F00668.jpg 1679 | ..\Traing_Data\images\CLIP069F00052.jpg 1680 | ..\Traing_Data\images\CLIP072F00922.jpg 1681 | ..\Traing_Data\images\CLIP068F00070.jpg 1682 | ..\Traing_Data\images\CLIP010004.jpg 1683 | ..\Traing_Data\images\2743.jpg 1684 | ..\Traing_Data\images\CLIP003063.jpg 1685 | ..\Traing_Data\images\CLIP071F00186.jpg 1686 | ..\Traing_Data\images\2805.jpg 1687 | ..\Traing_Data\images\CLIP066F00061.jpg 1688 | ..\Traing_Data\images\CLIP073F00633.jpg 1689 | ..\Traing_Data\images\CLIP006917.jpg 1690 | ..\Traing_Data\images\CLIP077F01874.jpg 1691 | ..\Traing_Data\images\CLIP077F00506.jpg 1692 | ..\Traing_Data\images\CLIP001368.jpg 1693 | ..\Traing_Data\images\CLIP066F00021.jpg 1694 | ..\Traing_Data\images\CLIP002519.jpg 1695 | ..\Traing_Data\images\CLIP007931.jpg 1696 | ..\Traing_Data\images\CLIP013456.jpg 1697 | ..\Traing_Data\images\CLIP072F00601.jpg 1698 | ..\Traing_Data\images\CLIP007975.jpg 1699 | ..\Traing_Data\images\4047.jpg 1700 | ..\Traing_Data\images\CLIP001378.jpg 1701 | ..\Traing_Data\images\CLIP003883.jpg 1702 | ..\Traing_Data\images\1997.jpg 1703 | ..\Traing_Data\images\CLIP077F00493.jpg 1704 | ..\Traing_Data\images\CLIP079F00420.jpg 1705 | ..\Traing_Data\images\CLIP012456.jpg 1706 | ..\Traing_Data\images\1868.jpg 1707 | ..\Traing_Data\images\CLIP073F01147.jpg 1708 | ..\Traing_Data\images\CLIP000628.jpg 1709 | ..\Traing_Data\images\CLIP001806.jpg 1710 | ..\Traing_Data\images\CLIP008654.jpg 1711 | ..\Traing_Data\images\CLIP001019.jpg 1712 | ..\Traing_Data\images\CLIP009720.jpg 1713 | ..\Traing_Data\images\CLIP001961.jpg 1714 | ..\Traing_Data\images\CLIP071F00165.jpg 1715 | ..\Traing_Data\images\CLIP001624.jpg 1716 | ..\Traing_Data\images\CLIP071F00193.jpg 1717 | ..\Traing_Data\images\2194.jpg 1718 | ..\Traing_Data\images\CLIP006498.jpg 1719 | ..\Traing_Data\images\CLIP064F00237.jpg 1720 | ..\Traing_Data\images\CLIP074F00502.jpg 1721 | ..\Traing_Data\images\CLIP008024.jpg 1722 | ..\Traing_Data\images\CLIP010783.jpg 1723 | ..\Traing_Data\images\3317.jpg 1724 | ..\Traing_Data\images\CLIP079F00419.jpg 1725 | ..\Traing_Data\images\CLIP072F00592.jpg 1726 | ..\Traing_Data\images\2337.jpg 1727 | ..\Traing_Data\images\2733.jpg 1728 | ..\Traing_Data\images\CLIP073F00662.jpg 1729 | ..\Traing_Data\images\CLIP012865.jpg 1730 | ..\Traing_Data\images\CLIP077F00636.jpg 1731 | ..\Traing_Data\images\CLIP004277.jpg 1732 | ..\Traing_Data\images\CLIP009824.jpg 1733 | ..\Traing_Data\images\2443.jpg 1734 | ..\Traing_Data\images\CLIP073F00635.jpg 1735 | ..\Traing_Data\images\CLIP002256.jpg 1736 | ..\Traing_Data\images\CLIP075F00596.jpg 1737 | ..\Traing_Data\images\CLIP008624.jpg 1738 | ..\Traing_Data\images\CLIP067F00029.jpg 1739 | ..\Traing_Data\images\CLIP003314.jpg 1740 | ..\Traing_Data\images\CLIP072F00930.jpg 1741 | ..\Traing_Data\images\CLIP071F00114.jpg 1742 | ..\Traing_Data\images\CLIP064F00398.jpg 1743 | ..\Traing_Data\images\CLIP072F00547.jpg 1744 | ..\Traing_Data\images\CLIP070F00041.jpg 1745 | ..\Traing_Data\images\CLIP068F00105.jpg 1746 | ..\Traing_Data\images\3933.jpg 1747 | ..\Traing_Data\images\3104.jpg 1748 | ..\Traing_Data\images\CLIP069F00171.jpg 1749 | ..\Traing_Data\images\3740.jpg 1750 | ..\Traing_Data\images\CLIP080F00495.jpg 1751 | ..\Traing_Data\images\CLIP074F00410.jpg 1752 | ..\Traing_Data\images\CLIP069F00080.jpg 1753 | ..\Traing_Data\images\CLIP005613.jpg 1754 | ..\Traing_Data\images\CLIP004412.jpg 1755 | ..\Traing_Data\images\CLIP080F00036.jpg 1756 | ..\Traing_Data\images\CLIP008830.jpg 1757 | ..\Traing_Data\images\CLIP010457.jpg 1758 | ..\Traing_Data\images\CLIP000868.jpg 1759 | ..\Traing_Data\images\CLIP073F00402.jpg 1760 | ..\Traing_Data\images\CLIP064F00232.jpg 1761 | ..\Traing_Data\images\2010.jpg 1762 | ..\Traing_Data\images\CLIP000882.jpg 1763 | ..\Traing_Data\images\CLIP068F00037.jpg 1764 | ..\Traing_Data\images\1431.jpg 1765 | ..\Traing_Data\images\CLIP012323.jpg 1766 | ..\Traing_Data\images\CLIP065F00003.jpg 1767 | ..\Traing_Data\images\3175.jpg 1768 | ..\Traing_Data\images\3633.jpg 1769 | ..\Traing_Data\images\CLIP001112.jpg 1770 | ..\Traing_Data\images\CLIP064F00337.jpg 1771 | ..\Traing_Data\images\CLIP008109.jpg 1772 | ..\Traing_Data\images\CLIP076F00328.jpg 1773 | ..\Traing_Data\images\2116.jpg 1774 | ..\Traing_Data\images\CLIP012017.jpg 1775 | ..\Traing_Data\images\CLIP074F00256.jpg 1776 | ..\Traing_Data\images\CLIP008231.jpg 1777 | ..\Traing_Data\images\CLIP072F00169.jpg 1778 | ..\Traing_Data\images\CLIP075F00042.jpg 1779 | ..\Traing_Data\images\CLIP069F00096.jpg 1780 | ..\Traing_Data\images\CLIP069F00181.jpg 1781 | ..\Traing_Data\images\CLIP003479.jpg 1782 | ..\Traing_Data\images\CLIP069F00178.jpg 1783 | ..\Traing_Data\images\CLIP066F00030.jpg 1784 | ..\Traing_Data\images\CLIP075F00532.jpg 1785 | ..\Traing_Data\images\CLIP068F00040.jpg 1786 | ..\Traing_Data\images\171.jpg 1787 | ..\Traing_Data\images\CLIP072F00174.jpg 1788 | ..\Traing_Data\images\1249.jpg 1789 | ..\Traing_Data\images\CLIP012971.jpg 1790 | ..\Traing_Data\images\CLIP074F00630.jpg 1791 | ..\Traing_Data\images\CLIP073F01034.jpg 1792 | ..\Traing_Data\images\CLIP079F00075.jpg 1793 | ..\Traing_Data\images\CLIP012143.jpg 1794 | ..\Traing_Data\images\CLIP006285.jpg 1795 | ..\Traing_Data\images\CLIP072F00487.jpg 1796 | ..\Traing_Data\images\248.jpg 1797 | ..\Traing_Data\images\CLIP006855.jpg 1798 | ..\Traing_Data\images\CLIP011523.jpg 1799 | ..\Traing_Data\images\CLIP074F00541.jpg 1800 | ..\Traing_Data\images\CLIP006051.jpg 1801 | ..\Traing_Data\images\CLIP006468.jpg 1802 | ..\Traing_Data\images\962.jpg 1803 | ..\Traing_Data\images\CLIP004211.jpg 1804 | ..\Traing_Data\images\CLIP079F00528.jpg 1805 | ..\Traing_Data\images\CLIP064F00151.jpg 1806 | ..\Traing_Data\images\CLIP069F00182.jpg 1807 | ..\Traing_Data\images\CLIP009438.jpg 1808 | ..\Traing_Data\images\CLIP068F00032.jpg 1809 | ..\Traing_Data\images\CLIP010968.jpg 1810 | ..\Traing_Data\images\CLIP011581.jpg 1811 | ..\Traing_Data\images\CLIP072F00061.jpg 1812 | ..\Traing_Data\images\CLIP077F00805.jpg 1813 | ..\Traing_Data\images\CLIP077F00335.jpg 1814 | ..\Traing_Data\images\CLIP066F00069.jpg 1815 | ..\Traing_Data\images\1381.jpg 1816 | ..\Traing_Data\images\CLIP072F00587.jpg 1817 | ..\Traing_Data\images\CLIP078F00503.jpg 1818 | ..\Traing_Data\images\CLIP072F00523.jpg 1819 | ..\Traing_Data\images\CLIP072F00504.jpg 1820 | ..\Traing_Data\images\99.jpg 1821 | ..\Traing_Data\images\CLIP010397.jpg 1822 | ..\Traing_Data\images\CLIP003037.jpg 1823 | ..\Traing_Data\images\CLIP071F00020.jpg 1824 | ..\Traing_Data\images\CLIP073F00800.jpg 1825 | ..\Traing_Data\images\CLIP073F00593.jpg 1826 | ..\Traing_Data\images\3883.jpg 1827 | ..\Traing_Data\images\CLIP073F00575.jpg 1828 | ..\Traing_Data\images\CLIP064F00200.jpg 1829 | ..\Traing_Data\images\CLIP074F00366.jpg 1830 | ..\Traing_Data\images\176.jpg 1831 | ..\Traing_Data\images\CLIP001249.jpg 1832 | ..\Traing_Data\images\CLIP072F00920.jpg 1833 | ..\Traing_Data\images\2351.jpg 1834 | ..\Traing_Data\images\CLIP072F00914.jpg 1835 | ..\Traing_Data\images\CLIP072F00301.jpg 1836 | ..\Traing_Data\images\CLIP077F01406.jpg 1837 | ..\Traing_Data\images\CLIP072F00688.jpg 1838 | ..\Traing_Data\images\CLIP011361.jpg 1839 | ..\Traing_Data\images\CLIP000830.jpg 1840 | ..\Traing_Data\images\CLIP067F00007.jpg 1841 | ..\Traing_Data\images\CLIP075F00455.jpg 1842 | ..\Traing_Data\images\CLIP069F00161.jpg 1843 | ..\Traing_Data\images\CLIP011457.jpg 1844 | ..\Traing_Data\images\CLIP072F00596.jpg 1845 | ..\Traing_Data\images\3488.jpg 1846 | ..\Traing_Data\images\CLIP072F00251.jpg 1847 | ..\Traing_Data\images\CLIP010296.jpg 1848 | ..\Traing_Data\images\CLIP001179.jpg 1849 | ..\Traing_Data\images\3660.jpg 1850 | ..\Traing_Data\images\1650.jpg 1851 | ..\Traing_Data\images\CLIP000748.jpg 1852 | ..\Traing_Data\images\CLIP007916.jpg 1853 | ..\Traing_Data\images\CLIP002170.jpg 1854 | ..\Traing_Data\images\CLIP011737.jpg 1855 | ..\Traing_Data\images\CLIP072F00321.jpg 1856 | ..\Traing_Data\images\CLIP074F00259.jpg 1857 | ..\Traing_Data\images\792.jpg 1858 | ..\Traing_Data\images\CLIP005452.jpg 1859 | ..\Traing_Data\images\CLIP066F00051.jpg 1860 | ..\Traing_Data\images\2231.jpg 1861 | ..\Traing_Data\images\CLIP074F00002.jpg 1862 | ..\Traing_Data\images\CLIP009026.jpg 1863 | ..\Traing_Data\images\CLIP071F00168.jpg 1864 | ..\Traing_Data\images\CLIP006488.jpg 1865 | ..\Traing_Data\images\106.jpg 1866 | ..\Traing_Data\images\CLIP007862.jpg 1867 | ..\Traing_Data\images\CLIP004618.jpg 1868 | ..\Traing_Data\images\CLIP002891.jpg 1869 | ..\Traing_Data\images\CLIP074F00205.jpg 1870 | ..\Traing_Data\images\CLIP069F00179.jpg 1871 | ..\Traing_Data\images\CLIP074F00556.jpg 1872 | ..\Traing_Data\images\CLIP003659.jpg 1873 | ..\Traing_Data\images\CLIP012923.jpg 1874 | ..\Traing_Data\images\4040.jpg 1875 | ..\Traing_Data\images\CLIP012740.jpg 1876 | ..\Traing_Data\images\392.jpg 1877 | ..\Traing_Data\images\CLIP076F00324.jpg 1878 | ..\Traing_Data\images\CLIP077F01407.jpg 1879 | ..\Traing_Data\images\CLIP000363.jpg 1880 | ..\Traing_Data\images\CLIP074F00270.jpg 1881 | ..\Traing_Data\images\CLIP004213.jpg 1882 | ..\Traing_Data\images\1448.jpg 1883 | ..\Traing_Data\images\CLIP064F00100.jpg 1884 | ..\Traing_Data\images\1151.jpg 1885 | ..\Traing_Data\images\CLIP076F00001.jpg 1886 | ..\Traing_Data\images\CLIP066F00018.jpg 1887 | ..\Traing_Data\images\CLIP005681.jpg 1888 | ..\Traing_Data\images\CLIP003345.jpg 1889 | ..\Traing_Data\images\CLIP072F00583.jpg 1890 | ..\Traing_Data\images\CLIP064F00061.jpg 1891 | ..\Traing_Data\images\CLIP003987.jpg 1892 | ..\Traing_Data\images\CLIP013825.jpg 1893 | ..\Traing_Data\images\CLIP073F00581.jpg 1894 | ..\Traing_Data\images\CLIP072F00082.jpg 1895 | ..\Traing_Data\images\CLIP014031.jpg 1896 | ..\Traing_Data\images\CLIP072F00552.jpg 1897 | ..\Traing_Data\images\CLIP065F00000.jpg 1898 | ..\Traing_Data\images\CLIP010756.jpg 1899 | ..\Traing_Data\images\CLIP073F00031.jpg 1900 | ..\Traing_Data\images\CLIP071F00278.jpg 1901 | ..\Traing_Data\images\CLIP008777.jpg 1902 | ..\Traing_Data\images\2520.jpg 1903 | ..\Traing_Data\images\CLIP013684.jpg 1904 | ..\Traing_Data\images\CLIP077F01299.jpg 1905 | ..\Traing_Data\images\CLIP070F00031.jpg 1906 | ..\Traing_Data\images\CLIP072F00584.jpg 1907 | ..\Traing_Data\images\CLIP069F00005.jpg 1908 | ..\Traing_Data\images\CLIP009396.jpg 1909 | ..\Traing_Data\images\CLIP074F00546.jpg 1910 | ..\Traing_Data\images\CLIP073F00579.jpg 1911 | ..\Traing_Data\images\CLIP011813.jpg 1912 | ..\Traing_Data\images\CLIP072F00932.jpg 1913 | ..\Traing_Data\images\CLIP010500.jpg 1914 | ..\Traing_Data\images\CLIP005630.jpg 1915 | ..\Traing_Data\images\CLIP074F00484.jpg 1916 | ..\Traing_Data\images\CLIP071F00286.jpg 1917 | ..\Traing_Data\images\CLIP075F00600.jpg 1918 | ..\Traing_Data\images\CLIP071F00249.jpg 1919 | ..\Traing_Data\images\CLIP072F00578.jpg 1920 | ..\Traing_Data\images\CLIP072F00702.jpg 1921 | ..\Traing_Data\images\CLIP073F01049.jpg 1922 | ..\Traing_Data\images\CLIP073F00611.jpg 1923 | ..\Traing_Data\images\CLIP080F00042.jpg 1924 | ..\Traing_Data\images\CLIP077F00332.jpg 1925 | ..\Traing_Data\images\641.jpg 1926 | ..\Traing_Data\images\CLIP012021.jpg 1927 | ..\Traing_Data\images\CLIP072F00083.jpg 1928 | ..\Traing_Data\images\CLIP005670.jpg 1929 | ..\Traing_Data\images\CLIP009921.jpg 1930 | ..\Traing_Data\images\CLIP004663.jpg 1931 | ..\Traing_Data\images\CLIP004236.jpg 1932 | ..\Traing_Data\images\2284.jpg 1933 | ..\Traing_Data\images\CLIP074F00730.jpg 1934 | ..\Traing_Data\images\CLIP077F00305.jpg 1935 | ..\Traing_Data\images\CLIP071F00247.jpg 1936 | ..\Traing_Data\images\CLIP013932.jpg 1937 | ..\Traing_Data\images\CLIP013349.jpg 1938 | ..\Traing_Data\images\CLIP075F00480.jpg 1939 | ..\Traing_Data\images\CLIP010995.jpg 1940 | ..\Traing_Data\images\CLIP067F00010.jpg 1941 | ..\Traing_Data\images\CLIP013891.jpg 1942 | ..\Traing_Data\images\CLIP074F00534.jpg 1943 | ..\Traing_Data\images\CLIP012270.jpg 1944 | ..\Traing_Data\images\CLIP004770.jpg 1945 | ..\Traing_Data\images\3217.jpg 1946 | ..\Traing_Data\images\CLIP072F00062.jpg 1947 | ..\Traing_Data\images\CLIP076F00041.jpg 1948 | ..\Traing_Data\images\CLIP008798.jpg 1949 | ..\Traing_Data\images\CLIP078F00702.jpg 1950 | ..\Traing_Data\images\CLIP006163.jpg 1951 | ..\Traing_Data\images\CLIP001401.jpg 1952 | ..\Traing_Data\images\CLIP072F00506.jpg 1953 | ..\Traing_Data\images\CLIP075F00519.jpg 1954 | ..\Traing_Data\images\CLIP009410.jpg 1955 | ..\Traing_Data\images\CLIP079F00547.jpg 1956 | ..\Traing_Data\images\CLIP008081.jpg 1957 | ..\Traing_Data\images\CLIP004740.jpg 1958 | ..\Traing_Data\images\336.jpg 1959 | ..\Traing_Data\images\CLIP004238.jpg 1960 | ..\Traing_Data\images\CLIP000257.jpg 1961 | ..\Traing_Data\images\CLIP005912.jpg 1962 | ..\Traing_Data\images\CLIP078F00507.jpg 1963 | ..\Traing_Data\images\CLIP072F00118.jpg 1964 | ..\Traing_Data\images\CLIP068F00004.jpg 1965 | ..\Traing_Data\images\CLIP010950.jpg 1966 | ..\Traing_Data\images\CLIP007142.jpg 1967 | ..\Traing_Data\images\CLIP073F00803.jpg 1968 | ..\Traing_Data\images\CLIP064F00164.jpg 1969 | ..\Traing_Data\images\674.jpg 1970 | ..\Traing_Data\images\CLIP006279.jpg 1971 | ..\Traing_Data\images\CLIP074F00526.jpg 1972 | ..\Traing_Data\images\CLIP011387.jpg 1973 | ..\Traing_Data\images\1804.jpg 1974 | ..\Traing_Data\images\CLIP012134.jpg 1975 | ..\Traing_Data\images\CLIP010357.jpg 1976 | ..\Traing_Data\images\CLIP005536.jpg 1977 | ..\Traing_Data\images\CLIP001337.jpg 1978 | ..\Traing_Data\images\CLIP011293.jpg 1979 | ..\Traing_Data\images\1894.jpg 1980 | ..\Traing_Data\images\CLIP075F01865.jpg 1981 | ..\Traing_Data\images\CLIP069F00095.jpg 1982 | ..\Traing_Data\images\CLIP012545.jpg 1983 | ..\Traing_Data\images\CLIP071F00201.jpg 1984 | ..\Traing_Data\images\CLIP002560.jpg 1985 | ..\Traing_Data\images\CLIP079F00656.jpg 1986 | ..\Traing_Data\images\CLIP068F00031.jpg 1987 | ..\Traing_Data\images\CLIP073F00176.jpg 1988 | ..\Traing_Data\images\CLIP004766.jpg 1989 | ..\Traing_Data\images\CLIP011912.jpg 1990 | ..\Traing_Data\images\CLIP073F00646.jpg 1991 | ..\Traing_Data\images\2493.jpg 1992 | ..\Traing_Data\images\971.jpg 1993 | ..\Traing_Data\images\CLIP073F00424.jpg 1994 | ..\Traing_Data\images\3497.jpg 1995 | ..\Traing_Data\images\CLIP000503.jpg 1996 | ..\Traing_Data\images\CLIP011521.jpg 1997 | ..\Traing_Data\images\CLIP066F00058.jpg 1998 | ..\Traing_Data\images\CLIP075F00598.jpg 1999 | ..\Traing_Data\images\CLIP006487.jpg 2000 | ..\Traing_Data\images\3279.jpg 2001 | ..\Traing_Data\images\CLIP006246.jpg 2002 | ..\Traing_Data\images\CLIP010670.jpg 2003 | ..\Traing_Data\images\CLIP069F00136.jpg 2004 | ..\Traing_Data\images\CLIP007234.jpg 2005 | ..\Traing_Data\images\CLIP008248.jpg 2006 | ..\Traing_Data\images\CLIP071F00240.jpg 2007 | ..\Traing_Data\images\CLIP011025.jpg 2008 | ..\Traing_Data\images\CLIP005365.jpg 2009 | ..\Traing_Data\images\CLIP011244.jpg 2010 | ..\Traing_Data\images\CLIP074F00303.jpg 2011 | ..\Traing_Data\images\CLIP064F00009.jpg 2012 | ..\Traing_Data\images\CLIP078F00970.jpg 2013 | ..\Traing_Data\images\CLIP072F00599.jpg 2014 | ..\Traing_Data\images\978.jpg 2015 | ..\Traing_Data\images\CLIP077F00514.jpg 2016 | ..\Traing_Data\images\CLIP011392.jpg 2017 | ..\Traing_Data\images\CLIP072F00327.jpg 2018 | ..\Traing_Data\images\CLIP012003.jpg 2019 | ..\Traing_Data\images\1467.jpg 2020 | ..\Traing_Data\images\CLIP068F00025.jpg 2021 | ..\Traing_Data\images\CLIP013390.jpg 2022 | ..\Traing_Data\images\CLIP078F00004.jpg 2023 | ..\Traing_Data\images\CLIP072F01025.jpg 2024 | ..\Traing_Data\images\CLIP072F00701.jpg 2025 | ..\Traing_Data\images\CLIP077F00505.jpg 2026 | ..\Traing_Data\images\CLIP075F01219.jpg 2027 | ..\Traing_Data\images\CLIP002746.jpg 2028 | ..\Traing_Data\images\CLIP005153.jpg 2029 | ..\Traing_Data\images\CLIP004555.jpg 2030 | ..\Traing_Data\images\CLIP004577.jpg 2031 | ..\Traing_Data\images\CLIP074F00496.jpg 2032 | ..\Traing_Data\images\CLIP073F00796.jpg 2033 | ..\Traing_Data\images\CLIP001925.jpg 2034 | ..\Traing_Data\images\CLIP010537.jpg 2035 | ..\Traing_Data\images\2985.jpg 2036 | ..\Traing_Data\images\CLIP006992.jpg 2037 | ..\Traing_Data\images\CLIP072F00227.jpg 2038 | ..\Traing_Data\images\CLIP005016.jpg 2039 | ..\Traing_Data\images\CLIP000057.jpg 2040 | ..\Traing_Data\images\CLIP072F00921.jpg 2041 | ..\Traing_Data\images\2258.jpg 2042 | ..\Traing_Data\images\CLIP001621.jpg 2043 | ..\Traing_Data\images\CLIP074F00209.jpg 2044 | ..\Traing_Data\images\CLIP012688.jpg 2045 | ..\Traing_Data\images\CLIP072F01075.jpg 2046 | ..\Traing_Data\images\CLIP004873.jpg 2047 | ..\Traing_Data\images\CLIP077F00341.jpg 2048 | ..\Traing_Data\images\CLIP009822.jpg 2049 | ..\Traing_Data\images\CLIP074F01077.jpg 2050 | ..\Traing_Data\images\CLIP073F00675.jpg 2051 | ..\Traing_Data\images\CLIP072F01007.jpg 2052 | ..\Traing_Data\images\CLIP008323.jpg 2053 | ..\Traing_Data\images\CLIP070F00065.jpg 2054 | ..\Traing_Data\images\CLIP073F00679.jpg 2055 | ..\Traing_Data\images\CLIP074F00533.jpg 2056 | ..\Traing_Data\images\CLIP066F00037.jpg 2057 | ..\Traing_Data\images\CLIP069F00056.jpg 2058 | ..\Traing_Data\images\CLIP066F00132.jpg 2059 | ..\Traing_Data\images\CLIP076F00659.jpg 2060 | ..\Traing_Data\images\CLIP007686.jpg 2061 | ..\Traing_Data\images\916.jpg 2062 | ..\Traing_Data\images\CLIP072F00855.jpg 2063 | ..\Traing_Data\images\CLIP011730.jpg 2064 | ..\Traing_Data\images\CLIP004254.jpg 2065 | ..\Traing_Data\images\CLIP000337.jpg 2066 | ..\Traing_Data\images\CLIP076F00301.jpg 2067 | ..\Traing_Data\images\CLIP005864.jpg 2068 | ..\Traing_Data\images\CLIP001651.jpg 2069 | ..\Traing_Data\images\CLIP001015.jpg 2070 | ..\Traing_Data\images\3148.jpg 2071 | ..\Traing_Data\images\CLIP007060.jpg 2072 | ..\Traing_Data\images\CLIP012351.jpg 2073 | ..\Traing_Data\images\657.jpg 2074 | ..\Traing_Data\images\CLIP072F00900.jpg 2075 | ..\Traing_Data\images\CLIP005549.jpg 2076 | ..\Traing_Data\images\2416.jpg 2077 | ..\Traing_Data\images\CLIP011939.jpg 2078 | ..\Traing_Data\images\CLIP064F00072.jpg 2079 | ..\Traing_Data\images\CLIP074F00532.jpg 2080 | ..\Traing_Data\images\CLIP004982.jpg 2081 | ..\Traing_Data\images\CLIP000145.jpg 2082 | ..\Traing_Data\images\CLIP072F00012.jpg 2083 | ..\Traing_Data\images\CLIP004566.jpg 2084 | ..\Traing_Data\images\CLIP013511.jpg 2085 | ..\Traing_Data\images\CLIP012598.jpg 2086 | ..\Traing_Data\images\CLIP010792.jpg 2087 | ..\Traing_Data\images\CLIP076F00038.jpg 2088 | ..\Traing_Data\images\CLIP073F00651.jpg 2089 | ..\Traing_Data\images\CLIP006402.jpg 2090 | ..\Traing_Data\images\CLIP064F00050.jpg 2091 | ..\Traing_Data\images\CLIP074F00538.jpg 2092 | ..\Traing_Data\images\CLIP072F01011.jpg 2093 | ..\Traing_Data\images\CLIP069F00050.jpg 2094 | ..\Traing_Data\images\CLIP009114.jpg 2095 | ..\Traing_Data\images\CLIP071F00234.jpg 2096 | ..\Traing_Data\images\CLIP008442.jpg 2097 | ..\Traing_Data\images\CLIP070F00037.jpg 2098 | ..\Traing_Data\images\CLIP074F00530.jpg 2099 | ..\Traing_Data\images\CLIP074F00269.jpg 2100 | ..\Traing_Data\images\CLIP005021.jpg 2101 | ..\Traing_Data\images\CLIP007534.jpg 2102 | ..\Traing_Data\images\CLIP077F00507.jpg 2103 | ..\Traing_Data\images\CLIP075F00511.jpg 2104 | ..\Traing_Data\images\CLIP003775.jpg 2105 | ..\Traing_Data\images\CLIP012705.jpg 2106 | ..\Traing_Data\images\CLIP010810.jpg 2107 | ..\Traing_Data\images\CLIP073F00785.jpg 2108 | ..\Traing_Data\images\CLIP010418.jpg 2109 | ..\Traing_Data\images\CLIP002326.jpg 2110 | ..\Traing_Data\images\CLIP071F00214.jpg 2111 | ..\Traing_Data\images\CLIP006415.jpg 2112 | ..\Traing_Data\images\CLIP072F00574.jpg 2113 | ..\Traing_Data\images\CLIP002691.jpg 2114 | ..\Traing_Data\images\CLIP073F00684.jpg 2115 | ..\Traing_Data\images\CLIP073F00625.jpg 2116 | ..\Traing_Data\images\CLIP012368.jpg 2117 | ..\Traing_Data\images\CLIP002960.jpg 2118 | ..\Traing_Data\images\CLIP006091.jpg 2119 | ..\Traing_Data\images\CLIP010848.jpg 2120 | ..\Traing_Data\images\CLIP075F00501.jpg 2121 | ..\Traing_Data\images\2425.jpg 2122 | ..\Traing_Data\images\CLIP073F00005.jpg 2123 | ..\Traing_Data\images\CLIP006569.jpg 2124 | ..\Traing_Data\images\3491.jpg 2125 | ..\Traing_Data\images\CLIP073F00037.jpg 2126 | ..\Traing_Data\images\CLIP067F00013.jpg 2127 | ..\Traing_Data\images\49.jpg 2128 | ..\Traing_Data\images\CLIP011874.jpg 2129 | ..\Traing_Data\images\CLIP071F00140.jpg 2130 | ..\Traing_Data\images\CLIP007340.jpg 2131 | ..\Traing_Data\images\CLIP069F00116.jpg 2132 | ..\Traing_Data\images\CLIP072F00050.jpg 2133 | ..\Traing_Data\images\CLIP075F00547.jpg 2134 | ..\Traing_Data\images\CLIP069F00027.jpg 2135 | ..\Traing_Data\images\CLIP000395.jpg 2136 | ..\Traing_Data\images\3334.jpg 2137 | ..\Traing_Data\images\CLIP074F00290.jpg 2138 | ..\Traing_Data\images\CLIP079F00005.jpg 2139 | ..\Traing_Data\images\CLIP003990.jpg 2140 | ..\Traing_Data\images\CLIP000761.jpg 2141 | ..\Traing_Data\images\CLIP012952.jpg 2142 | ..\Traing_Data\images\3342.jpg 2143 | ..\Traing_Data\images\CLIP069F00140.jpg 2144 | ..\Traing_Data\images\CLIP071F00151.jpg 2145 | ..\Traing_Data\images\CLIP003071.jpg 2146 | ..\Traing_Data\images\CLIP069F00070.jpg 2147 | ..\Traing_Data\images\CLIP077F01409.jpg 2148 | ..\Traing_Data\images\CLIP008945.jpg 2149 | ..\Traing_Data\images\CLIP077F00808.jpg 2150 | ..\Traing_Data\images\CLIP078F00971.jpg 2151 | ..\Traing_Data\images\CLIP075F00467.jpg 2152 | ..\Traing_Data\images\CLIP077F00497.jpg 2153 | ..\Traing_Data\images\CLIP009687.jpg 2154 | ..\Traing_Data\images\CLIP071F00248.jpg 2155 | ..\Traing_Data\images\CLIP012408.jpg 2156 | ..\Traing_Data\images\CLIP069F00109.jpg 2157 | ..\Traing_Data\images\CLIP008076.jpg 2158 | ..\Traing_Data\images\CLIP075F00744.jpg 2159 | ..\Traing_Data\images\CLIP080F00039.jpg 2160 | ..\Traing_Data\images\CLIP073F00801.jpg 2161 | ..\Traing_Data\images\CLIP073F00589.jpg 2162 | ..\Traing_Data\images\CLIP008168.jpg 2163 | ..\Traing_Data\images\CLIP069F00134.jpg 2164 | ..\Traing_Data\images\CLIP076F00004.jpg 2165 | ..\Traing_Data\images\CLIP002420.jpg 2166 | ..\Traing_Data\images\CLIP073F00884.jpg 2167 | ..\Traing_Data\images\CLIP072F00298.jpg 2168 | ..\Traing_Data\images\CLIP012849.jpg 2169 | ..\Traing_Data\images\CLIP071F00191.jpg 2170 | ..\Traing_Data\images\CLIP072F00139.jpg 2171 | ..\Traing_Data\images\CLIP078F00982.jpg 2172 | ..\Traing_Data\images\CLIP072F00926.jpg 2173 | ..\Traing_Data\images\CLIP076F00658.jpg 2174 | ..\Traing_Data\images\CLIP072F00078.jpg 2175 | ..\Traing_Data\images\CLIP072F00772.jpg 2176 | ..\Traing_Data\images\CLIP064F00369.jpg 2177 | ..\Traing_Data\images\CLIP074F00170.jpg 2178 | ..\Traing_Data\images\CLIP003067.jpg 2179 | ..\Traing_Data\images\CLIP069F00107.jpg 2180 | ..\Traing_Data\images\CLIP068F00082.jpg 2181 | ..\Traing_Data\images\CLIP008157.jpg 2182 | ..\Traing_Data\images\CLIP000517.jpg 2183 | ..\Traing_Data\images\1730.jpg 2184 | ..\Traing_Data\images\CLIP005077.jpg 2185 | ..\Traing_Data\images\CLIP075F00481.jpg 2186 | ..\Traing_Data\images\CLIP066F00063.jpg 2187 | ..\Traing_Data\images\3038.jpg 2188 | ..\Traing_Data\images\CLIP065F00009.jpg 2189 | ..\Traing_Data\images\CLIP072F00228.jpg 2190 | ..\Traing_Data\images\CLIP072F00597.jpg 2191 | ..\Traing_Data\images\CLIP071F00133.jpg 2192 | ..\Traing_Data\images\CLIP075F00503.jpg 2193 | ..\Traing_Data\images\CLIP005160.jpg 2194 | ..\Traing_Data\images\CLIP001491.jpg 2195 | ..\Traing_Data\images\CLIP073F01220.jpg 2196 | ..\Traing_Data\images\CLIP005050.jpg 2197 | ..\Traing_Data\images\CLIP001801.jpg 2198 | ..\Traing_Data\images\CLIP071F00121.jpg 2199 | ..\Traing_Data\images\CLIP006115.jpg 2200 | ..\Traing_Data\images\575.jpg 2201 | ..\Traing_Data\images\CLIP013092.jpg 2202 | ..\Traing_Data\images\CLIP012005.jpg 2203 | ..\Traing_Data\images\CLIP001317.jpg 2204 | ..\Traing_Data\images\CLIP064F00159.jpg 2205 | ..\Traing_Data\images\CLIP072F00607.jpg 2206 | ..\Traing_Data\images\CLIP010172.jpg 2207 | ..\Traing_Data\images\CLIP066F00029.jpg 2208 | ..\Traing_Data\images\CLIP011097.jpg 2209 | ..\Traing_Data\images\CLIP073F01262.jpg 2210 | ..\Traing_Data\images\CLIP005455.jpg 2211 | ..\Traing_Data\images\CLIP072F00043.jpg 2212 | ..\Traing_Data\images\CLIP069F00081.jpg 2213 | ..\Traing_Data\images\CLIP075F01213.jpg 2214 | ..\Traing_Data\images\CLIP074F01074.jpg 2215 | ..\Traing_Data\images\CLIP073F00217.jpg 2216 | ..\Traing_Data\images\CLIP001935.jpg 2217 | ..\Traing_Data\images\CLIP007549.jpg 2218 | ..\Traing_Data\images\553.jpg 2219 | ..\Traing_Data\images\CLIP077F00804.jpg 2220 | ..\Traing_Data\images\890.jpg 2221 | ..\Traing_Data\images\2013.jpg 2222 | ..\Traing_Data\images\CLIP012089.jpg 2223 | ..\Traing_Data\images\CLIP005119.jpg 2224 | ..\Traing_Data\images\CLIP008028.jpg 2225 | ..\Traing_Data\images\CLIP064F00250.jpg 2226 | ..\Traing_Data\images\CLIP002813.jpg 2227 | ..\Traing_Data\images\CLIP073F00792.jpg 2228 | ..\Traing_Data\images\CLIP013280.jpg 2229 | ..\Traing_Data\images\CLIP003425.jpg 2230 | ..\Traing_Data\images\CLIP072F00303.jpg 2231 | ..\Traing_Data\images\CLIP071F00292.jpg 2232 | ..\Traing_Data\images\CLIP074F00499.jpg 2233 | ..\Traing_Data\images\CLIP064F00226.jpg 2234 | ..\Traing_Data\images\CLIP007097.jpg 2235 | ..\Traing_Data\images\CLIP001850.jpg 2236 | ..\Traing_Data\images\CLIP072F00106.jpg 2237 | ..\Traing_Data\images\1878.jpg 2238 | ..\Traing_Data\images\CLIP072F00317.jpg 2239 | ..\Traing_Data\images\41.jpg 2240 | ..\Traing_Data\images\CLIP064F00363.jpg 2241 | ..\Traing_Data\images\636.jpg 2242 | ..\Traing_Data\images\CLIP072F00059.jpg 2243 | ..\Traing_Data\images\CLIP072F00127.jpg 2244 | ..\Traing_Data\images\CLIP000975.jpg 2245 | ..\Traing_Data\images\CLIP010251.jpg 2246 | ..\Traing_Data\images\CLIP001213.jpg 2247 | ..\Traing_Data\images\CLIP013361.jpg 2248 | ..\Traing_Data\images\CLIP011601.jpg 2249 | ..\Traing_Data\images\CLIP000625.jpg 2250 | ..\Traing_Data\images\CLIP070F00044.jpg 2251 | ..\Traing_Data\images\CLIP071F00258.jpg 2252 | ..\Traing_Data\images\CLIP079F00456.jpg 2253 | ..\Traing_Data\images\CLIP004114.jpg 2254 | ..\Traing_Data\images\CLIP009832.jpg 2255 | ..\Traing_Data\images\CLIP073F00900.jpg 2256 | ..\Traing_Data\images\2960.jpg 2257 | ..\Traing_Data\images\CLIP010085.jpg 2258 | ..\Traing_Data\images\1823.jpg 2259 | ..\Traing_Data\images\2779.jpg 2260 | ..\Traing_Data\images\CLIP002663.jpg 2261 | ..\Traing_Data\images\CLIP072F00257.jpg 2262 | ..\Traing_Data\images\CLIP075F00050.jpg 2263 | ..\Traing_Data\images\CLIP006864.jpg 2264 | ..\Traing_Data\images\CLIP073F00177.jpg 2265 | ..\Traing_Data\images\CLIP003820.jpg 2266 | ..\Traing_Data\images\CLIP008240.jpg 2267 | ..\Traing_Data\images\2424.jpg 2268 | ..\Traing_Data\images\CLIP001017.jpg 2269 | ..\Traing_Data\images\CLIP079F00421.jpg 2270 | ..\Traing_Data\images\CLIP069F00126.jpg 2271 | ..\Traing_Data\images\CLIP009099.jpg 2272 | ..\Traing_Data\images\CLIP071F00264.jpg 2273 | ..\Traing_Data\images\CLIP064F00043.jpg 2274 | ..\Traing_Data\images\90.jpg 2275 | ..\Traing_Data\images\CLIP012685.jpg 2276 | ..\Traing_Data\images\CLIP072F00196.jpg 2277 | ..\Traing_Data\images\CLIP069F00009.jpg 2278 | ..\Traing_Data\images\CLIP012303.jpg 2279 | ..\Traing_Data\images\CLIP070F00019.jpg 2280 | ..\Traing_Data\images\CLIP071F00175.jpg 2281 | ..\Traing_Data\images\CLIP071F00088.jpg 2282 | ..\Traing_Data\images\CLIP012902.jpg 2283 | ..\Traing_Data\images\CLIP069F00053.jpg 2284 | ..\Traing_Data\images\CLIP074F01075.jpg 2285 | ..\Traing_Data\images\2564.jpg 2286 | ..\Traing_Data\images\CLIP072F00334.jpg 2287 | ..\Traing_Data\images\CLIP010256.jpg 2288 | ..\Traing_Data\images\CLIP007001.jpg 2289 | ..\Traing_Data\images\CLIP064F00254.jpg 2290 | ..\Traing_Data\images\CLIP073F00668.jpg 2291 | ..\Traing_Data\images\CLIP009054.jpg 2292 | ..\Traing_Data\images\CLIP078F00981.jpg 2293 | ..\Traing_Data\images\CLIP008023.jpg 2294 | ..\Traing_Data\images\CLIP074F00206.jpg 2295 | ..\Traing_Data\images\CLIP073F00062.jpg 2296 | ..\Traing_Data\images\CLIP072F00170.jpg 2297 | ..\Traing_Data\images\CLIP071F00017.jpg 2298 | ..\Traing_Data\images\CLIP064F00094.jpg 2299 | ..\Traing_Data\images\CLIP078F00504.jpg 2300 | ..\Traing_Data\images\CLIP009309.jpg 2301 | ..\Traing_Data\images\CLIP070F00050.jpg 2302 | ..\Traing_Data\images\CLIP009188.jpg 2303 | ..\Traing_Data\images\CLIP075F00508.jpg 2304 | ..\Traing_Data\images\CLIP074F00267.jpg 2305 | ..\Traing_Data\images\CLIP069F00168.jpg 2306 | ..\Traing_Data\images\CLIP008900.jpg 2307 | ..\Traing_Data\images\1434.jpg 2308 | ..\Traing_Data\images\CLIP007444.jpg 2309 | ..\Traing_Data\images\CLIP075F01625.jpg 2310 | ..\Traing_Data\images\CLIP008690.jpg 2311 | ..\Traing_Data\images\CLIP077F00301.jpg 2312 | ..\Traing_Data\images\CLIP068F00081.jpg 2313 | ..\Traing_Data\images\CLIP075F00756.jpg 2314 | ..\Traing_Data\images\CLIP071F00202.jpg 2315 | ..\Traing_Data\images\CLIP002132.jpg 2316 | ..\Traing_Data\images\CLIP073F00601.jpg 2317 | ..\Traing_Data\images\CLIP000436.jpg 2318 | ..\Traing_Data\images\CLIP012327.jpg 2319 | ..\Traing_Data\images\CLIP069F00018.jpg 2320 | ..\Traing_Data\images\CLIP074F01063.jpg 2321 | ..\Traing_Data\images\CLIP000634.jpg 2322 | ..\Traing_Data\images\3077.jpg 2323 | ..\Traing_Data\images\CLIP071F00217.jpg 2324 | ..\Traing_Data\images\CLIP073F00682.jpg 2325 | ..\Traing_Data\images\CLIP071F00224.jpg 2326 | ..\Traing_Data\images\CLIP000215.jpg 2327 | ..\Traing_Data\images\CLIP077F00509.jpg 2328 | ..\Traing_Data\images\CLIP013867.jpg 2329 | ..\Traing_Data\images\CLIP006201.jpg 2330 | ..\Traing_Data\images\907.jpg 2331 | ..\Traing_Data\images\CLIP012297.jpg 2332 | ..\Traing_Data\images\CLIP074F00614.jpg 2333 | ..\Traing_Data\images\CLIP002727.jpg 2334 | ..\Traing_Data\images\CLIP075F00745.jpg 2335 | ..\Traing_Data\images\CLIP071F00236.jpg 2336 | ..\Traing_Data\images\CLIP072F00689.jpg 2337 | ..\Traing_Data\images\CLIP073F00599.jpg 2338 | ..\Traing_Data\images\391.jpg 2339 | ..\Traing_Data\images\CLIP011118.jpg 2340 | ..\Traing_Data\images\2217.jpg 2341 | ..\Traing_Data\images\4067.jpg 2342 | ..\Traing_Data\images\CLIP072F00104.jpg 2343 | ..\Traing_Data\images\CLIP006919.jpg 2344 | ..\Traing_Data\images\CLIP072F01074.jpg 2345 | ..\Traing_Data\images\CLIP071F00120.jpg 2346 | ..\Traing_Data\images\CLIP068F00044.jpg 2347 | ..\Traing_Data\images\CLIP074F00626.jpg 2348 | ..\Traing_Data\images\CLIP011548.jpg 2349 | ..\Traing_Data\images\CLIP072F00593.jpg 2350 | ..\Traing_Data\images\CLIP075F00425.jpg 2351 | ..\Traing_Data\images\CLIP072F01008.jpg 2352 | ..\Traing_Data\images\CLIP072F00870.jpg 2353 | ..\Traing_Data\images\CLIP008575.jpg 2354 | ..\Traing_Data\images\CLIP072F00577.jpg 2355 | ..\Traing_Data\images\CLIP010154.jpg 2356 | ..\Traing_Data\images\CLIP080F00002.jpg 2357 | ..\Traing_Data\images\CLIP072F00906.jpg 2358 | ..\Traing_Data\images\CLIP006221.jpg 2359 | ..\Traing_Data\images\CLIP009807.jpg 2360 | ..\Traing_Data\images\CLIP072F00230.jpg 2361 | ..\Traing_Data\images\CLIP072F00693.jpg 2362 | ..\Traing_Data\images\CLIP074F00293.jpg 2363 | ..\Traing_Data\images\CLIP072F00912.jpg 2364 | ..\Traing_Data\images\CLIP014035.jpg 2365 | ..\Traing_Data\images\1766.jpg 2366 | ..\Traing_Data\images\CLIP012335.jpg 2367 | ..\Traing_Data\images\CLIP078F03121.jpg 2368 | ..\Traing_Data\images\CLIP075F00465.jpg 2369 | ..\Traing_Data\images\CLIP000134.jpg 2370 | ..\Traing_Data\images\CLIP074F00462.jpg 2371 | ..\Traing_Data\images\CLIP002637.jpg 2372 | ..\Traing_Data\images\CLIP006853.jpg 2373 | ..\Traing_Data\images\CLIP002781.jpg 2374 | ..\Traing_Data\images\CLIP003237.jpg 2375 | ..\Traing_Data\images\CLIP007253.jpg 2376 | ..\Traing_Data\images\CLIP002610.jpg 2377 | ..\Traing_Data\images\CLIP075F00667.jpg 2378 | ..\Traing_Data\images\CLIP007664.jpg 2379 | ..\Traing_Data\images\CLIP072F01017.jpg 2380 | ..\Traing_Data\images\3983.jpg 2381 | ..\Traing_Data\images\CLIP073F01032.jpg 2382 | ..\Traing_Data\images\CLIP078F00125.jpg 2383 | ..\Traing_Data\images\CLIP014089.jpg 2384 | ..\Traing_Data\images\CLIP072F00697.jpg 2385 | ..\Traing_Data\images\CLIP071F00272.jpg 2386 | ..\Traing_Data\images\CLIP071F00131.jpg 2387 | ..\Traing_Data\images\CLIP007442.jpg 2388 | ..\Traing_Data\images\CLIP071F00276.jpg 2389 | ..\Traing_Data\images\CLIP069F00023.jpg 2390 | ..\Traing_Data\images\CLIP069F00165.jpg 2391 | ..\Traing_Data\images\CLIP001739.jpg 2392 | ..\Traing_Data\images\CLIP072F00760.jpg 2393 | ..\Traing_Data\images\CLIP078F00983.jpg 2394 | ..\Traing_Data\images\CLIP010592.jpg 2395 | ..\Traing_Data\images\CLIP073F00626.jpg 2396 | ..\Traing_Data\images\CLIP070F00062.jpg 2397 | ..\Traing_Data\images\CLIP078F00500.jpg 2398 | ..\Traing_Data\images\CLIP072F01024.jpg 2399 | ..\Traing_Data\images\617.jpg 2400 | ..\Traing_Data\images\CLIP075F00479.jpg 2401 | ..\Traing_Data\images\CLIP010459.jpg 2402 | ..\Traing_Data\images\CLIP000958.jpg 2403 | ..\Traing_Data\images\CLIP006017.jpg 2404 | ..\Traing_Data\images\CLIP072F01006.jpg 2405 | ..\Traing_Data\images\CLIP072F00582.jpg 2406 | ..\Traing_Data\images\CLIP075F00476.jpg 2407 | ..\Traing_Data\images\CLIP074F00477.jpg 2408 | ..\Traing_Data\images\CLIP076F00303.jpg 2409 | ..\Traing_Data\images\CLIP072F01072.jpg 2410 | ..\Traing_Data\images\CLIP003459.jpg 2411 | ..\Traing_Data\images\CLIP069F00187.jpg 2412 | ..\Traing_Data\images\1490.jpg 2413 | ..\Traing_Data\images\2658.jpg 2414 | ..\Traing_Data\images\CLIP075F00755.jpg 2415 | ..\Traing_Data\images\CLIP001040.jpg 2416 | ..\Traing_Data\images\CLIP075F00372.jpg 2417 | ..\Traing_Data\images\CLIP074F00603.jpg 2418 | ..\Traing_Data\images\CLIP072F00919.jpg 2419 | ..\Traing_Data\images\CLIP072F00873.jpg 2420 | ..\Traing_Data\images\CLIP013109.jpg 2421 | ..\Traing_Data\images\CLIP069F00185.jpg 2422 | ..\Traing_Data\images\CLIP072F00322.jpg 2423 | ..\Traing_Data\images\2705.jpg 2424 | ..\Traing_Data\images\CLIP069F00145.jpg 2425 | ..\Traing_Data\images\CLIP002580.jpg 2426 | ..\Traing_Data\images\CLIP072F01071.jpg 2427 | ..\Traing_Data\images\CLIP007040.jpg 2428 | ..\Traing_Data\images\CLIP067F00025.jpg 2429 | ..\Traing_Data\images\CLIP075F02106.jpg 2430 | ..\Traing_Data\images\CLIP069F00007.jpg 2431 | ..\Traing_Data\images\CLIP006513.jpg 2432 | ..\Traing_Data\images\CLIP001397.jpg 2433 | ..\Traing_Data\images\CLIP012985.jpg 2434 | ..\Traing_Data\images\CLIP008628.jpg 2435 | ..\Traing_Data\images\CLIP008994.jpg 2436 | ..\Traing_Data\images\CLIP011418.jpg 2437 | ..\Traing_Data\images\CLIP071F00206.jpg 2438 | ..\Traing_Data\images\CLIP003098.jpg 2439 | ..\Traing_Data\images\1569.jpg 2440 | ..\Traing_Data\images\CLIP075F00672.jpg 2441 | ..\Traing_Data\images\CLIP077F01492.jpg 2442 | ..\Traing_Data\images\CLIP002096.jpg 2443 | ..\Traing_Data\images\CLIP076F00006.jpg 2444 | ..\Traing_Data\images\CLIP009168.jpg 2445 | ..\Traing_Data\images\CLIP069F00089.jpg 2446 | ..\Traing_Data\images\CLIP006142.jpg 2447 | ..\Traing_Data\images\CLIP009760.jpg 2448 | ..\Traing_Data\images\CLIP004174.jpg 2449 | ..\Traing_Data\images\CLIP003572.jpg 2450 | ..\Traing_Data\images\CLIP001966.jpg 2451 | ..\Traing_Data\images\CLIP002225.jpg 2452 | ..\Traing_Data\images\CLIP002840.jpg 2453 | ..\Traing_Data\images\CLIP072F00939.jpg 2454 | ..\Traing_Data\images\3512.jpg 2455 | ..\Traing_Data\images\CLIP013163.jpg 2456 | ..\Traing_Data\images\CLIP013888.jpg 2457 | ..\Traing_Data\images\CLIP066F00054.jpg 2458 | ..\Traing_Data\images\CLIP003412.jpg 2459 | ..\Traing_Data\images\CLIP073F00607.jpg 2460 | ..\Traing_Data\images\CLIP009539.jpg 2461 | ..\Traing_Data\images\CLIP076F00011.jpg 2462 | ..\Traing_Data\images\CLIP005787.jpg 2463 | ..\Traing_Data\images\CLIP073F00666.jpg 2464 | ..\Traing_Data\images\1050.jpg 2465 | ..\Traing_Data\images\1111.jpg 2466 | ..\Traing_Data\images\CLIP012402.jpg 2467 | ..\Traing_Data\images\CLIP073F00043.jpg 2468 | ..\Traing_Data\images\CLIP002426.jpg 2469 | ..\Traing_Data\images\CLIP007519.jpg 2470 | ..\Traing_Data\images\CLIP010962.jpg 2471 | ..\Traing_Data\images\54.jpg 2472 | ..\Traing_Data\images\CLIP072F00135.jpg 2473 | ..\Traing_Data\images\CLIP005446.jpg 2474 | ..\Traing_Data\images\CLIP072F00143.jpg 2475 | ..\Traing_Data\images\CLIP002784.jpg 2476 | ..\Traing_Data\images\CLIP072F00195.jpg 2477 | ..\Traing_Data\images\CLIP069F00184.jpg 2478 | ..\Traing_Data\images\CLIP013659.jpg 2479 | ..\Traing_Data\images\CLIP076F00003.jpg 2480 | ..\Traing_Data\images\CLIP075F00540.jpg 2481 | ..\Traing_Data\images\CLIP071F00256.jpg 2482 | ..\Traing_Data\images\CLIP067F00031.jpg 2483 | ..\Traing_Data\images\CLIP008092.jpg 2484 | ..\Traing_Data\images\CLIP013794.jpg 2485 | ..\Traing_Data\images\CLIP068F00030.jpg 2486 | ..\Traing_Data\images\CLIP072F00089.jpg 2487 | ..\Traing_Data\images\CLIP068F00066.jpg 2488 | ..\Traing_Data\images\CLIP074F00529.jpg 2489 | ..\Traing_Data\images\CLIP071F00078.jpg 2490 | ..\Traing_Data\images\CLIP008040.jpg 2491 | ..\Traing_Data\images\CLIP008279.jpg 2492 | ..\Traing_Data\images\CLIP010233.jpg 2493 | ..\Traing_Data\images\CLIP073F00182.jpg 2494 | ..\Traing_Data\images\CLIP072F01019.jpg 2495 | ..\Traing_Data\images\CLIP064F00099.jpg 2496 | ..\Traing_Data\images\CLIP074F01065.jpg 2497 | ..\Traing_Data\images\1.jpg 2498 | ..\Traing_Data\images\CLIP003498.jpg 2499 | ..\Traing_Data\images\CLIP071F00279.jpg 2500 | ..\Traing_Data\images\CLIP066F00070.jpg 2501 | ..\Traing_Data\images\CLIP011861.jpg 2502 | ..\Traing_Data\images\3009.jpg 2503 | ..\Traing_Data\images\CLIP009531.jpg 2504 | ..\Traing_Data\images\CLIP009078.jpg 2505 | ..\Traing_Data\images\CLIP009603.jpg 2506 | ..\Traing_Data\images\CLIP079F00417.jpg 2507 | ..\Traing_Data\images\CLIP013310.jpg 2508 | ..\Traing_Data\images\CLIP071F00139.jpg 2509 | ..\Traing_Data\images\CLIP005602.jpg 2510 | ..\Traing_Data\images\1993.jpg 2511 | ..\Traing_Data\images\CLIP074F00623.jpg 2512 | ..\Traing_Data\images\CLIP077F00508.jpg 2513 | ..\Traing_Data\images\1753.jpg 2514 | ..\Traing_Data\images\125.jpg 2515 | ..\Traing_Data\images\CLIP071F00184.jpg 2516 | ..\Traing_Data\images\CLIP005679.jpg 2517 | ..\Traing_Data\images\2597.jpg 2518 | ..\Traing_Data\images\CLIP075F00368.jpg 2519 | ..\Traing_Data\images\CLIP064F00395.jpg 2520 | ..\Traing_Data\images\3931.jpg 2521 | ..\Traing_Data\images\CLIP006558.jpg 2522 | ..\Traing_Data\images\CLIP071F00008.jpg 2523 | ..\Traing_Data\images\CLIP072F00783.jpg 2524 | ..\Traing_Data\images\CLIP004121.jpg 2525 | ..\Traing_Data\images\CLIP008504.jpg 2526 | ..\Traing_Data\images\CLIP006379.jpg 2527 | ..\Traing_Data\images\CLIP012657.jpg 2528 | ..\Traing_Data\images\CLIP073F00645.jpg 2529 | ..\Traing_Data\images\CLIP078F00976.jpg 2530 | ..\Traing_Data\images\CLIP009552.jpg 2531 | ..\Traing_Data\images\CLIP000204.jpg 2532 | ..\Traing_Data\images\CLIP072F00931.jpg 2533 | ..\Traing_Data\images\CLIP072F00576.jpg 2534 | ..\Traing_Data\images\667.jpg 2535 | ..\Traing_Data\images\CLIP013802.jpg 2536 | ..\Traing_Data\images\CLIP000520.jpg 2537 | ..\Traing_Data\images\CLIP072F00241.jpg 2538 | ..\Traing_Data\images\CLIP070F00056.jpg 2539 | ..\Traing_Data\images\CLIP072F00002.jpg 2540 | ..\Traing_Data\images\CLIP075F00545.jpg 2541 | ..\Traing_Data\images\999.jpg 2542 | ..\Traing_Data\images\CLIP064F00396.jpg 2543 | ..\Traing_Data\images\CLIP073F00600.jpg 2544 | ..\Traing_Data\images\CLIP074F00609.jpg 2545 | ..\Traing_Data\images\CLIP010482.jpg 2546 | ..\Traing_Data\images\1095.jpg 2547 | ..\Traing_Data\images\CLIP064F00077.jpg 2548 | ..\Traing_Data\images\CLIP010331.jpg 2549 | ..\Traing_Data\images\CLIP069F00176.jpg 2550 | ..\Traing_Data\images\CLIP072F00161.jpg 2551 | ..\Traing_Data\images\CLIP009343.jpg 2552 | ..\Traing_Data\images\CLIP072F00097.jpg 2553 | ..\Traing_Data\images\CLIP005545.jpg 2554 | ..\Traing_Data\images\CLIP013906.jpg 2555 | ..\Traing_Data\images\CLIP005969.jpg 2556 | ..\Traing_Data\images\CLIP077F01494.jpg 2557 | ..\Traing_Data\images\CLIP072F00117.jpg 2558 | ..\Traing_Data\images\CLIP076F00039.jpg 2559 | ..\Traing_Data\images\CLIP012393.jpg 2560 | ..\Traing_Data\images\2618.jpg 2561 | ..\Traing_Data\images\CLIP073F00676.jpg 2562 | ..\Traing_Data\images\CLIP012409.jpg 2563 | ..\Traing_Data\images\CLIP074F00260.jpg 2564 | ..\Traing_Data\images\3625.jpg 2565 | ..\Traing_Data\images\CLIP013600.jpg 2566 | ..\Traing_Data\images\CLIP077F00300.jpg 2567 | ..\Traing_Data\images\3529.jpg 2568 | ..\Traing_Data\images\CLIP064F00382.jpg 2569 | ..\Traing_Data\images\CLIP007481.jpg 2570 | ..\Traing_Data\images\CLIP065F00002.jpg 2571 | ..\Traing_Data\images\CLIP009662.jpg 2572 | ..\Traing_Data\images\CLIP005937.jpg 2573 | ..\Traing_Data\images\CLIP071F00220.jpg 2574 | ..\Traing_Data\images\CLIP010327.jpg 2575 | ..\Traing_Data\images\CLIP007674.jpg 2576 | ..\Traing_Data\images\CLIP070F00020.jpg 2577 | ..\Traing_Data\images\CLIP071F00313.jpg 2578 | ..\Traing_Data\images\3004.jpg 2579 | ..\Traing_Data\images\CLIP000233.jpg 2580 | ..\Traing_Data\images\CLIP072F00187.jpg 2581 | ..\Traing_Data\images\CLIP072F00328.jpg 2582 | ..\Traing_Data\images\CLIP074F00213.jpg 2583 | ..\Traing_Data\images\CLIP066F00017.jpg 2584 | ..\Traing_Data\images\CLIP077F01300.jpg 2585 | ..\Traing_Data\images\CLIP011379.jpg 2586 | ..\Traing_Data\images\CLIP010054.jpg 2587 | ..\Traing_Data\images\CLIP009468.jpg 2588 | ..\Traing_Data\images\CLIP008591.jpg 2589 | ..\Traing_Data\images\CLIP072F00307.jpg 2590 | ..\Traing_Data\images\CLIP069F00060.jpg 2591 | ..\Traing_Data\images\CLIP064F00171.jpg 2592 | ..\Traing_Data\images\3957.jpg 2593 | ..\Traing_Data\images\CLIP071F00312.jpg 2594 | ..\Traing_Data\images\CLIP072F00028.jpg 2595 | ..\Traing_Data\images\CLIP008127.jpg 2596 | ..\Traing_Data\images\1188.jpg 2597 | ..\Traing_Data\images\CLIP072F00903.jpg 2598 | ..\Traing_Data\images\CLIP066F00046.jpg 2599 | ..\Traing_Data\images\CLIP075F00603.jpg 2600 | ..\Traing_Data\images\CLIP012793.jpg 2601 | ..\Traing_Data\images\CLIP014002.jpg 2602 | ..\Traing_Data\images\CLIP072F00570.jpg 2603 | ..\Traing_Data\images\CLIP006188.jpg 2604 | ..\Traing_Data\images\CLIP013950.jpg 2605 | ..\Traing_Data\images\CLIP074F00281.jpg 2606 | ..\Traing_Data\images\CLIP073F00414.jpg 2607 | ..\Traing_Data\images\CLIP075F00747.jpg 2608 | ..\Traing_Data\images\CLIP002071.jpg 2609 | ..\Traing_Data\images\CLIP069F00094.jpg 2610 | ..\Traing_Data\images\CLIP068F00057.jpg 2611 | ..\Traing_Data\images\CLIP004042.jpg 2612 | ..\Traing_Data\images\2498.jpg 2613 | ..\Traing_Data\images\CLIP013510.jpg 2614 | ..\Traing_Data\images\CLIP072F00522.jpg 2615 | ..\Traing_Data\images\CLIP066F00039.jpg 2616 | ..\Traing_Data\images\CLIP075F01001.jpg 2617 | ..\Traing_Data\images\359.jpg 2618 | ..\Traing_Data\images\CLIP013072.jpg 2619 | ..\Traing_Data\images\3130.jpg 2620 | ..\Traing_Data\images\CLIP007690.jpg 2621 | ..\Traing_Data\images\CLIP075F00051.jpg 2622 | ..\Traing_Data\images\1711.jpg 2623 | ..\Traing_Data\images\CLIP071F00110.jpg 2624 | ..\Traing_Data\images\CLIP073F00285.jpg 2625 | ..\Traing_Data\images\3100.jpg 2626 | ..\Traing_Data\images\CLIP072F00096.jpg 2627 | ..\Traing_Data\images\CLIP003404.jpg 2628 | ..\Traing_Data\images\CLIP072F00242.jpg 2629 | ..\Traing_Data\images\3654.jpg 2630 | ..\Traing_Data\images\CLIP002920.jpg 2631 | ..\Traing_Data\images\CLIP072F00110.jpg 2632 | ..\Traing_Data\images\CLIP074F00365.jpg 2633 | ..\Traing_Data\images\CLIP073F00184.jpg 2634 | ..\Traing_Data\images\CLIP006287.jpg 2635 | ..\Traing_Data\images\CLIP009008.jpg 2636 | ..\Traing_Data\images\CLIP006761.jpg 2637 | ..\Traing_Data\images\CLIP068F00000.jpg 2638 | ..\Traing_Data\images\CLIP073F01310.jpg 2639 | ..\Traing_Data\images\CLIP010958.jpg 2640 | ..\Traing_Data\images\CLIP068F00026.jpg 2641 | ..\Traing_Data\images\CLIP077F00806.jpg 2642 | ..\Traing_Data\images\CLIP064F00216.jpg 2643 | ..\Traing_Data\images\CLIP000020.jpg 2644 | ..\Traing_Data\images\CLIP074F00629.jpg 2645 | ..\Traing_Data\images\CLIP000093.jpg 2646 | ..\Traing_Data\images\CLIP075F00041.jpg 2647 | ..\Traing_Data\images\CLIP072F00915.jpg 2648 | ..\Traing_Data\images\CLIP068F00093.jpg 2649 | ..\Traing_Data\images\CLIP073F00639.jpg 2650 | ..\Traing_Data\images\CLIP071F00261.jpg 2651 | ..\Traing_Data\images\CLIP012006.jpg 2652 | ..\Traing_Data\images\CLIP000845.jpg 2653 | ..\Traing_Data\images\CLIP071F00162.jpg 2654 | ..\Traing_Data\images\CLIP005254.jpg 2655 | ..\Traing_Data\images\CLIP003315.jpg 2656 | ..\Traing_Data\images\CLIP064F00134.jpg 2657 | ..\Traing_Data\images\CLIP079F00074.jpg 2658 | ..\Traing_Data\images\CLIP073F00069.jpg 2659 | ..\Traing_Data\images\CLIP069F00025.jpg 2660 | ..\Traing_Data\images\CLIP075F00470.jpg 2661 | ..\Traing_Data\images\CLIP071F00300.jpg 2662 | ..\Traing_Data\images\CLIP008193.jpg 2663 | ..\Traing_Data\images\2128.jpg 2664 | ..\Traing_Data\images\CLIP064F00372.jpg 2665 | ..\Traing_Data\images\CLIP065F00013.jpg 2666 | ..\Traing_Data\images\CLIP011787.jpg 2667 | ..\Traing_Data\images\CLIP009525.jpg 2668 | ..\Traing_Data\images\CLIP007884.jpg 2669 | ..\Traing_Data\images\CLIP001882.jpg 2670 | ..\Traing_Data\images\2716.jpg 2671 | ..\Traing_Data\images\CLIP010232.jpg 2672 | ..\Traing_Data\images\CLIP064F00123.jpg 2673 | ..\Traing_Data\images\CLIP005905.jpg 2674 | ..\Traing_Data\images\1637.jpg 2675 | ..\Traing_Data\images\CLIP074F00272.jpg 2676 | ..\Traing_Data\images\CLIP008414.jpg 2677 | ..\Traing_Data\images\CLIP072F00243.jpg 2678 | ..\Traing_Data\images\CLIP071F00137.jpg 2679 | ..\Traing_Data\images\1383.jpg 2680 | ..\Traing_Data\images\CLIP012063.jpg 2681 | ..\Traing_Data\images\CLIP077F00306.jpg 2682 | ..\Traing_Data\images\2672.jpg 2683 | ..\Traing_Data\images\CLIP010114.jpg 2684 | ..\Traing_Data\images\CLIP075F00673.jpg 2685 | ..\Traing_Data\images\CLIP071F00091.jpg 2686 | ..\Traing_Data\images\CLIP006239.jpg 2687 | ..\Traing_Data\images\3328.jpg 2688 | ..\Traing_Data\images\CLIP007135.jpg 2689 | ..\Traing_Data\images\1988.jpg 2690 | ..\Traing_Data\images\CLIP071F00087.jpg 2691 | ..\Traing_Data\images\CLIP010708.jpg 2692 | ..\Traing_Data\images\CLIP074F00301.jpg 2693 | ..\Traing_Data\images\CLIP072F00492.jpg 2694 | ..\Traing_Data\images\CLIP079F00661.jpg 2695 | ..\Traing_Data\images\CLIP000018.jpg 2696 | ..\Traing_Data\images\215.jpg 2697 | ..\Traing_Data\images\CLIP070F00066.jpg 2698 | ..\Traing_Data\images\CLIP076F00289.jpg 2699 | ..\Traing_Data\images\CLIP003460.jpg 2700 | ..\Traing_Data\images\CLIP009738.jpg 2701 | ..\Traing_Data\images\CLIP009372.jpg 2702 | ..\Traing_Data\images\CLIP072F00567.jpg 2703 | ..\Traing_Data\images\CLIP008541.jpg 2704 | ..\Traing_Data\images\CLIP007704.jpg 2705 | ..\Traing_Data\images\CLIP071F00158.jpg 2706 | ..\Traing_Data\images\CLIP074F00624.jpg 2707 | ..\Traing_Data\images\CLIP075F00427.jpg 2708 | ..\Traing_Data\images\CLIP012249.jpg 2709 | ..\Traing_Data\images\CLIP004160.jpg 2710 | ..\Traing_Data\images\CLIP077F00634.jpg 2711 | ..\Traing_Data\images\CLIP006803.jpg 2712 | ..\Traing_Data\images\CLIP072F00137.jpg 2713 | ..\Traing_Data\images\CLIP003607.jpg 2714 | ..\Traing_Data\images\CLIP002686.jpg 2715 | ..\Traing_Data\images\CLIP066F00041.jpg 2716 | ..\Traing_Data\images\CLIP064F00400.jpg 2717 | ..\Traing_Data\images\CLIP078F00129.jpg 2718 | ..\Traing_Data\images\2023.jpg 2719 | ..\Traing_Data\images\CLIP007478.jpg 2720 | ..\Traing_Data\images\CLIP069F00043.jpg 2721 | ..\Traing_Data\images\CLIP064F00385.jpg 2722 | ..\Traing_Data\images\CLIP068F00080.jpg 2723 | ..\Traing_Data\images\CLIP004524.jpg 2724 | ..\Traing_Data\images\CLIP073F00034.jpg 2725 | ..\Traing_Data\images\CLIP071F00253.jpg 2726 | ..\Traing_Data\images\3420.jpg 2727 | ..\Traing_Data\images\CLIP071F00233.jpg 2728 | ..\Traing_Data\images\CLIP003571.jpg 2729 | ..\Traing_Data\images\CLIP073F00661.jpg 2730 | ..\Traing_Data\images\CLIP003909.jpg 2731 | ..\Traing_Data\images\CLIP077F00342.jpg 2732 | ..\Traing_Data\images\2222.jpg 2733 | ..\Traing_Data\images\CLIP068F00011.jpg 2734 | ..\Traing_Data\images\CLIP000285.jpg 2735 | ..\Traing_Data\images\CLIP072F00947.jpg 2736 | ..\Traing_Data\images\CLIP009611.jpg 2737 | ..\Traing_Data\images\CLIP002487.jpg 2738 | ..\Traing_Data\images\CLIP001402.jpg 2739 | ..\Traing_Data\images\CLIP009418.jpg 2740 | ..\Traing_Data\images\CLIP011316.jpg 2741 | ..\Traing_Data\images\CLIP077F00000.jpg 2742 | ..\Traing_Data\images\CLIP068F00094.jpg 2743 | ..\Traing_Data\images\CLIP067F00019.jpg 2744 | ..\Traing_Data\images\CLIP004552.jpg 2745 | ..\Traing_Data\images\CLIP004320.jpg 2746 | ..\Traing_Data\images\CLIP075F00999.jpg 2747 | ..\Traing_Data\images\3911.jpg 2748 | ..\Traing_Data\images\3301.jpg 2749 | ..\Traing_Data\images\CLIP002037.jpg 2750 | ..\Traing_Data\images\CLIP009024.jpg 2751 | ..\Traing_Data\images\CLIP079F01730.jpg 2752 | ..\Traing_Data\images\CLIP008437.jpg 2753 | ..\Traing_Data\images\CLIP071F00118.jpg 2754 | ..\Traing_Data\images\CLIP074F00500.jpg 2755 | ..\Traing_Data\images\CLIP066F00028.jpg 2756 | ..\Traing_Data\images\CLIP069F00011.jpg 2757 | ..\Traing_Data\images\CLIP071F00094.jpg 2758 | ..\Traing_Data\images\CLIP079F00418.jpg 2759 | ..\Traing_Data\images\CLIP071F00211.jpg 2760 | ..\Traing_Data\images\CLIP072F00936.jpg 2761 | ..\Traing_Data\images\CLIP004515.jpg 2762 | ..\Traing_Data\images\CLIP009253.jpg 2763 | ..\Traing_Data\images\CLIP073F00624.jpg 2764 | ..\Traing_Data\images\CLIP004682.jpg 2765 | ..\Traing_Data\images\CLIP009050.jpg 2766 | ..\Traing_Data\images\CLIP008567.jpg 2767 | ..\Traing_Data\images\CLIP007533.jpg 2768 | ..\Traing_Data\images\CLIP008720.jpg 2769 | ..\Traing_Data\images\CLIP073F00291.jpg 2770 | ..\Traing_Data\images\CLIP006076.jpg 2771 | ..\Traing_Data\images\CLIP003312.jpg 2772 | ..\Traing_Data\images\CLIP072F00254.jpg 2773 | ..\Traing_Data\images\CLIP071F00288.jpg 2774 | ..\Traing_Data\images\CLIP071F00160.jpg 2775 | ..\Traing_Data\images\CLIP069F00163.jpg 2776 | ..\Traing_Data\images\CLIP075F00451.jpg 2777 | ..\Traing_Data\images\CLIP013881.jpg 2778 | ..\Traing_Data\images\CLIP073F00438.jpg 2779 | ..\Traing_Data\images\CLIP003568.jpg 2780 | ..\Traing_Data\images\CLIP073F01145.jpg 2781 | ..\Traing_Data\images\CLIP001803.jpg 2782 | ..\Traing_Data\images\CLIP077F00333.jpg 2783 | ..\Traing_Data\images\CLIP010780.jpg 2784 | ..\Traing_Data\images\CLIP006170.jpg 2785 | ..\Traing_Data\images\CLIP002049.jpg 2786 | ..\Traing_Data\images\CLIP074F00412.jpg 2787 | ..\Traing_Data\images\CLIP067F00016.jpg 2788 | ..\Traing_Data\images\CLIP002387.jpg 2789 | ..\Traing_Data\images\CLIP005416.jpg 2790 | ..\Traing_Data\images\CLIP073F00578.jpg 2791 | ..\Traing_Data\images\CLIP074F00493.jpg 2792 | ..\Traing_Data\images\CLIP005768.jpg 2793 | ..\Traing_Data\images\CLIP007369.jpg 2794 | ..\Traing_Data\images\CLIP066F00034.jpg 2795 | ..\Traing_Data\images\CLIP065F00004.jpg 2796 | ..\Traing_Data\images\CLIP072F00149.jpg 2797 | ..\Traing_Data\images\CLIP006976.jpg 2798 | ..\Traing_Data\images\1522.jpg 2799 | ..\Traing_Data\images\CLIP007815.jpg 2800 | ..\Traing_Data\images\CLIP072F00052.jpg 2801 | ..\Traing_Data\images\CLIP005721.jpg 2802 | ..\Traing_Data\images\60.jpg 2803 | ..\Traing_Data\images\CLIP002917.jpg 2804 | ..\Traing_Data\images\CLIP074F00369.jpg 2805 | ..\Traing_Data\images\CLIP072F00129.jpg 2806 | ..\Traing_Data\images\CLIP013583.jpg 2807 | ..\Traing_Data\images\CLIP072F00560.jpg 2808 | ..\Traing_Data\images\CLIP069F00068.jpg 2809 | ..\Traing_Data\images\CLIP007963.jpg 2810 | ..\Traing_Data\images\3678.jpg 2811 | ..\Traing_Data\images\CLIP012934.jpg 2812 | ..\Traing_Data\images\CLIP003210.jpg 2813 | ..\Traing_Data\images\CLIP072F00095.jpg 2814 | ..\Traing_Data\images\CLIP068F00027.jpg 2815 | ..\Traing_Data\images\CLIP003965.jpg 2816 | ..\Traing_Data\images\CLIP013100.jpg 2817 | ..\Traing_Data\images\CLIP074F00456.jpg 2818 | ..\Traing_Data\images\CLIP009528.jpg 2819 | ..\Traing_Data\images\CLIP007550.jpg 2820 | ..\Traing_Data\images\CLIP078F00979.jpg 2821 | ..\Traing_Data\images\CLIP072F00125.jpg 2822 | ..\Traing_Data\images\CLIP001257.jpg 2823 | ..\Traing_Data\images\CLIP071F00221.jpg 2824 | ..\Traing_Data\images\1779.jpg 2825 | ..\Traing_Data\images\CLIP005944.jpg 2826 | ..\Traing_Data\images\1655.jpg 2827 | ..\Traing_Data\images\CLIP077F00485.jpg 2828 | ..\Traing_Data\images\CLIP075F00589.jpg 2829 | ..\Traing_Data\images\CLIP013849.jpg 2830 | ..\Traing_Data\images\CLIP001407.jpg 2831 | ..\Traing_Data\images\CLIP012999.jpg 2832 | ..\Traing_Data\images\CLIP008619.jpg 2833 | ..\Traing_Data\images\3395.jpg 2834 | ..\Traing_Data\images\CLIP013027.jpg 2835 | ..\Traing_Data\images\CLIP073F00401.jpg 2836 | ..\Traing_Data\images\CLIP005687.jpg 2837 | ..\Traing_Data\images\CLIP000968.jpg 2838 | ..\Traing_Data\images\CLIP008734.jpg 2839 | ..\Traing_Data\images\CLIP009367.jpg 2840 | ..\Traing_Data\images\CLIP068F00029.jpg 2841 | ..\Traing_Data\images\CLIP012261.jpg 2842 | ..\Traing_Data\images\CLIP005667.jpg 2843 | ..\Traing_Data\images\CLIP006362.jpg 2844 | ..\Traing_Data\images\CLIP077F00480.jpg 2845 | ..\Traing_Data\images\CLIP012561.jpg 2846 | ..\Traing_Data\images\CLIP073F00629.jpg 2847 | ..\Traing_Data\images\602.jpg 2848 | ..\Traing_Data\images\CLIP071F00280.jpg 2849 | ..\Traing_Data\images\CLIP073F00793.jpg 2850 | ..\Traing_Data\images\CLIP072F00029.jpg 2851 | ..\Traing_Data\images\CLIP064F00346.jpg 2852 | ..\Traing_Data\images\CLIP073F00879.jpg 2853 | ..\Traing_Data\images\CLIP012635.jpg 2854 | ..\Traing_Data\images\CLIP013459.jpg 2855 | ..\Traing_Data\images\CLIP069F00131.jpg 2856 | ..\Traing_Data\images\CLIP001335.jpg 2857 | ..\Traing_Data\images\CLIP006199.jpg 2858 | ..\Traing_Data\images\CLIP069F00108.jpg 2859 | ..\Traing_Data\images\CLIP075F00366.jpg 2860 | ..\Traing_Data\images\CLIP012597.jpg 2861 | ..\Traing_Data\images\CLIP006371.jpg 2862 | ..\Traing_Data\images\CLIP075F00800.jpg 2863 | ..\Traing_Data\images\CLIP078F00974.jpg 2864 | ..\Traing_Data\images\CLIP076F00300.jpg 2865 | ..\Traing_Data\images\CLIP002503.jpg 2866 | ..\Traing_Data\images\CLIP064F00034.jpg 2867 | ..\Traing_Data\images\622.jpg 2868 | ..\Traing_Data\images\CLIP064F00163.jpg 2869 | ..\Traing_Data\images\CLIP066F00138.jpg 2870 | ..\Traing_Data\images\CLIP074F00300.jpg 2871 | ..\Traing_Data\images\CLIP071F00174.jpg 2872 | ..\Traing_Data\images\CLIP064F00242.jpg 2873 | ..\Traing_Data\images\CLIP072F00588.jpg 2874 | ..\Traing_Data\images\CLIP076F00290.jpg 2875 | ..\Traing_Data\images\CLIP001677.jpg 2876 | ..\Traing_Data\images\CLIP000142.jpg 2877 | ..\Traing_Data\images\2480.jpg 2878 | ..\Traing_Data\images\836.jpg 2879 | ..\Traing_Data\images\CLIP002573.jpg 2880 | ..\Traing_Data\images\CLIP006366.jpg 2881 | ..\Traing_Data\images\CLIP002400.jpg 2882 | ..\Traing_Data\images\CLIP073F00878.jpg 2883 | ..\Traing_Data\images\CLIP064F00348.jpg 2884 | ..\Traing_Data\images\2342.jpg 2885 | ..\Traing_Data\images\1683.jpg 2886 | ..\Traing_Data\images\CLIP006429.jpg 2887 | ..\Traing_Data\images\CLIP004646.jpg 2888 | ..\Traing_Data\images\CLIP072F00496.jpg 2889 | ..\Traing_Data\images\200.jpg 2890 | ..\Traing_Data\images\CLIP011916.jpg 2891 | ..\Traing_Data\images\2024.jpg 2892 | ..\Traing_Data\images\CLIP003247.jpg 2893 | ..\Traing_Data\images\CLIP011635.jpg 2894 | ..\Traing_Data\images\CLIP004693.jpg 2895 | ..\Traing_Data\images\CLIP000466.jpg 2896 | ..\Traing_Data\images\CLIP072F00524.jpg 2897 | ..\Traing_Data\images\CLIP070F00072.jpg 2898 | ..\Traing_Data\images\CLIP079F00073.jpg 2899 | ..\Traing_Data\images\CLIP074F00295.jpg 2900 | ..\Traing_Data\images\CLIP000449.jpg 2901 | ..\Traing_Data\images\2274.jpg 2902 | ..\Traing_Data\images\CLIP072F00554.jpg 2903 | ..\Traing_Data\images\CLIP005722.jpg 2904 | ..\Traing_Data\images\CLIP072F00595.jpg 2905 | ..\Traing_Data\images\CLIP079F01731.jpg 2906 | ..\Traing_Data\images\CLIP011713.jpg 2907 | ..\Traing_Data\images\CLIP072F00538.jpg 2908 | ..\Traing_Data\images\CLIP072F00167.jpg 2909 | ..\Traing_Data\images\CLIP071F00002.jpg 2910 | ..\Traing_Data\images\CLIP064F00071.jpg 2911 | ..\Traing_Data\images\CLIP071F00014.jpg 2912 | ..\Traing_Data\images\CLIP007922.jpg 2913 | ..\Traing_Data\images\CLIP002490.jpg 2914 | ..\Traing_Data\images\CLIP071F00194.jpg 2915 | ..\Traing_Data\images\CLIP006066.jpg 2916 | ..\Traing_Data\images\CLIP064F00082.jpg 2917 | ..\Traing_Data\images\CLIP077F00482.jpg 2918 | ..\Traing_Data\images\CLIP076F00288.jpg 2919 | ..\Traing_Data\images\CLIP070F00061.jpg 2920 | ..\Traing_Data\images\159.jpg 2921 | ..\Traing_Data\images\CLIP075F00363.jpg 2922 | ..\Traing_Data\images\CLIP012812.jpg 2923 | ..\Traing_Data\images\CLIP074F00368.jpg 2924 | ..\Traing_Data\images\CLIP073F01014.jpg 2925 | ..\Traing_Data\images\CLIP070F00070.jpg 2926 | ..\Traing_Data\images\CLIP008960.jpg 2927 | ..\Traing_Data\images\CLIP008527.jpg 2928 | ..\Traing_Data\images\CLIP072F00600.jpg 2929 | ..\Traing_Data\images\CLIP008305.jpg 2930 | ..\Traing_Data\images\CLIP072F00103.jpg 2931 | ..\Traing_Data\images\CLIP073F00649.jpg 2932 | ..\Traing_Data\images\CLIP071F00275.jpg 2933 | ..\Traing_Data\images\CLIP007422.jpg 2934 | ..\Traing_Data\images\CLIP077F01563.jpg 2935 | ..\Traing_Data\images\CLIP009779.jpg 2936 | ..\Traing_Data\images\CLIP076F00316.jpg 2937 | ..\Traing_Data\images\CLIP005101.jpg 2938 | ..\Traing_Data\images\CLIP003483.jpg 2939 | ..\Traing_Data\images\CLIP008677.jpg 2940 | ..\Traing_Data\images\CLIP069F00114.jpg 2941 | ..\Traing_Data\images\2952.jpg 2942 | ..\Traing_Data\images\CLIP079F00523.jpg 2943 | ..\Traing_Data\images\CLIP073F00216.jpg 2944 | ..\Traing_Data\images\CLIP072F00759.jpg 2945 | ..\Traing_Data\images\CLIP074F00480.jpg 2946 | ..\Traing_Data\images\CLIP004719.jpg 2947 | ..\Traing_Data\images\CLIP003012.jpg 2948 | ..\Traing_Data\images\CLIP068F00014.jpg 2949 | ..\Traing_Data\images\CLIP071F00309.jpg 2950 | ..\Traing_Data\images\CLIP073F00004.jpg 2951 | ..\Traing_Data\images\CLIP069F00120.jpg 2952 | ..\Traing_Data\images\CLIP072F00690.jpg 2953 | ..\Traing_Data\images\CLIP010129.jpg 2954 | ..\Traing_Data\images\CLIP008484.jpg 2955 | ..\Traing_Data\images\CLIP069F00175.jpg 2956 | ..\Traing_Data\images\CLIP071F00173.jpg 2957 | ..\Traing_Data\images\CLIP064F00404.jpg 2958 | ..\Traing_Data\images\CLIP069F00033.jpg 2959 | ..\Traing_Data\images\CLIP004968.jpg 2960 | ..\Traing_Data\images\CLIP008447.jpg 2961 | ..\Traing_Data\images\CLIP072F00016.jpg 2962 | ..\Traing_Data\images\CLIP011164.jpg 2963 | ..\Traing_Data\images\CLIP073F00643.jpg 2964 | ..\Traing_Data\images\CLIP000133.jpg 2965 | ..\Traing_Data\images\CLIP072F00606.jpg 2966 | ..\Traing_Data\images\1914.jpg 2967 | ..\Traing_Data\images\CLIP001398.jpg 2968 | ..\Traing_Data\images\1064.jpg 2969 | ..\Traing_Data\images\CLIP011909.jpg 2970 | ..\Traing_Data\images\CLIP069F00039.jpg 2971 | ..\Traing_Data\images\CLIP000990.jpg 2972 | ..\Traing_Data\images\CLIP006785.jpg 2973 | ..\Traing_Data\images\1787.jpg 2974 | ..\Traing_Data\images\CLIP012512.jpg 2975 | ..\Traing_Data\images\CLIP001535.jpg 2976 | ..\Traing_Data\images\CLIP072F00526.jpg 2977 | ..\Traing_Data\images\CLIP079F01732.jpg 2978 | ..\Traing_Data\images\CLIP001350.jpg 2979 | ..\Traing_Data\images\CLIP069F00074.jpg 2980 | ..\Traing_Data\images\3458.jpg 2981 | ..\Traing_Data\images\CLIP077F01869.jpg 2982 | ..\Traing_Data\images\CLIP002255.jpg 2983 | ..\Traing_Data\images\CLIP078F00984.jpg 2984 | ..\Traing_Data\images\CLIP000621.jpg 2985 | ..\Traing_Data\images\CLIP009043.jpg 2986 | ..\Traing_Data\images\CLIP072F00240.jpg 2987 | ..\Traing_Data\images\CLIP005038.jpg 2988 | ..\Traing_Data\images\3465.jpg 2989 | ..\Traing_Data\images\CLIP075F00453.jpg 2990 | ..\Traing_Data\images\CLIP073F00640.jpg 2991 | ..\Traing_Data\images\CLIP069F00154.jpg 2992 | ..\Traing_Data\images\CLIP074F00463.jpg 2993 | ..\Traing_Data\images\CLIP005266.jpg 2994 | ..\Traing_Data\images\CLIP073F00885.jpg 2995 | ..\Traing_Data\images\CLIP079F00515.jpg 2996 | ..\Traing_Data\images\CLIP072F00236.jpg 2997 | ..\Traing_Data\images\3242.jpg 2998 | ..\Traing_Data\images\CLIP075F00762.jpg 2999 | ..\Traing_Data\images\CLIP073F00437.jpg 3000 | ..\Traing_Data\images\CLIP064F00032.jpg 3001 | ..\Traing_Data\images\CLIP000446.jpg 3002 | ..\Traing_Data\images\CLIP064F00047.jpg 3003 | ..\Traing_Data\images\CLIP008851.jpg 3004 | ..\Traing_Data\images\CLIP013798.jpg 3005 | ..\Traing_Data\images\CLIP001018.jpg 3006 | ..\Traing_Data\images\CLIP075F00604.jpg 3007 | ..\Traing_Data\images\CLIP067F00020.jpg 3008 | ..\Traing_Data\images\CLIP069F00133.jpg 3009 | ..\Traing_Data\images\CLIP007230.jpg 3010 | ..\Traing_Data\images\CLIP072F00238.jpg 3011 | ..\Traing_Data\images\CLIP069F00017.jpg 3012 | ..\Traing_Data\images\CLIP001451.jpg 3013 | ..\Traing_Data\images\CLIP079F00658.jpg 3014 | ..\Traing_Data\images\CLIP074F00201.jpg 3015 | ..\Traing_Data\images\CLIP004182.jpg 3016 | ..\Traing_Data\images\2982.jpg 3017 | ..\Traing_Data\images\CLIP072F00512.jpg 3018 | ..\Traing_Data\images\CLIP010979.jpg 3019 | ..\Traing_Data\images\CLIP005273.jpg 3020 | ..\Traing_Data\images\CLIP070F00032.jpg 3021 | ..\Traing_Data\images\CLIP072F00586.jpg 3022 | ..\Traing_Data\images\2553.jpg 3023 | ..\Traing_Data\images\1356.jpg 3024 | ..\Traing_Data\images\CLIP069F00113.jpg 3025 | ..\Traing_Data\images\CLIP072F00314.jpg 3026 | ..\Traing_Data\images\CLIP011896.jpg 3027 | ..\Traing_Data\images\CLIP073F00901.jpg 3028 | ..\Traing_Data\images\CLIP073F00787.jpg 3029 | ..\Traing_Data\images\CLIP077F00302.jpg 3030 | ..\Traing_Data\images\CLIP078F00002.jpg 3031 | ..\Traing_Data\images\CLIP079F00522.jpg 3032 | ..\Traing_Data\images\CLIP001928.jpg 3033 | ..\Traing_Data\images\CLIP071F00267.jpg 3034 | ..\Traing_Data\images\CLIP004920.jpg 3035 | ..\Traing_Data\images\CLIP073F00791.jpg 3036 | ..\Traing_Data\images\CLIP005804.jpg 3037 | ..\Traing_Data\images\CLIP073F00218.jpg 3038 | ..\Traing_Data\images\CLIP002724.jpg 3039 | ..\Traing_Data\images\CLIP077F01403.jpg 3040 | ..\Traing_Data\images\CLIP072F00026.jpg 3041 | ..\Traing_Data\images\CLIP006317.jpg 3042 | ..\Traing_Data\images\CLIP010253.jpg 3043 | ..\Traing_Data\images\CLIP011948.jpg 3044 | ..\Traing_Data\images\603.jpg 3045 | ..\Traing_Data\images\12.jpg 3046 | ..\Traing_Data\images\CLIP074F00265.jpg 3047 | ..\Traing_Data\images\CLIP013785.jpg 3048 | ..\Traing_Data\images\CLIP075F00669.jpg 3049 | ..\Traing_Data\images\CLIP071F00228.jpg 3050 | ..\Traing_Data\images\CLIP069F00085.jpg 3051 | ..\Traing_Data\images\CLIP075F02104.jpg 3052 | ..\Traing_Data\images\CLIP001196.jpg 3053 | ..\Traing_Data\images\2338.jpg 3054 | ..\Traing_Data\images\CLIP072F00904.jpg 3055 | ..\Traing_Data\images\CLIP006231.jpg 3056 | ..\Traing_Data\images\CLIP070F00039.jpg 3057 | ..\Traing_Data\images\CLIP072F00181.jpg 3058 | ..\Traing_Data\images\CLIP008924.jpg 3059 | ..\Traing_Data\images\CLIP075F00521.jpg 3060 | ..\Traing_Data\images\CLIP010416.jpg 3061 | ..\Traing_Data\images\CLIP006024.jpg 3062 | ..\Traing_Data\images\CLIP004697.jpg 3063 | ..\Traing_Data\images\CLIP072F00959.jpg 3064 | ..\Traing_Data\images\CLIP069F00102.jpg 3065 | ..\Traing_Data\images\CLIP071F00100.jpg 3066 | ..\Traing_Data\images\CLIP001580.jpg 3067 | ..\Traing_Data\images\CLIP072F00532.jpg 3068 | ..\Traing_Data\images\CLIP071F00304.jpg 3069 | ..\Traing_Data\images\CLIP001098.jpg 3070 | ..\Traing_Data\images\CLIP014075.jpg 3071 | ..\Traing_Data\images\CLIP073F01261.jpg 3072 | ..\Traing_Data\images\CLIP071F00284.jpg 3073 | ..\Traing_Data\images\CLIP070F00017.jpg 3074 | ..\Traing_Data\images\CLIP008608.jpg 3075 | ..\Traing_Data\images\CLIP069F00004.jpg 3076 | ..\Traing_Data\images\CLIP010077.jpg 3077 | ..\Traing_Data\images\CLIP005897.jpg 3078 | ..\Traing_Data\images\CLIP075F00530.jpg 3079 | ..\Traing_Data\images\CLIP010517.jpg 3080 | ..\Traing_Data\images\CLIP013860.jpg 3081 | ..\Traing_Data\images\CLIP064F00063.jpg 3082 | ..\Traing_Data\images\CLIP072F00329.jpg 3083 | ..\Traing_Data\images\30.jpg 3084 | ..\Traing_Data\images\CLIP000207.jpg 3085 | ..\Traing_Data\images\CLIP011849.jpg 3086 | ..\Traing_Data\images\3274.jpg 3087 | ..\Traing_Data\images\CLIP072F00263.jpg 3088 | ..\Traing_Data\images\CLIP071F00170.jpg 3089 | ..\Traing_Data\images\CLIP080F00454.jpg 3090 | ..\Traing_Data\images\CLIP006247.jpg 3091 | ..\Traing_Data\images\CLIP067F00033.jpg 3092 | ..\Traing_Data\images\CLIP072F00509.jpg 3093 | ..\Traing_Data\images\CLIP076F00285.jpg 3094 | ..\Traing_Data\images\CLIP064F00324.jpg 3095 | ..\Traing_Data\images\1706.jpg 3096 | ..\Traing_Data\images\CLIP073F00427.jpg 3097 | ..\Traing_Data\images\CLIP071F00301.jpg 3098 | ..\Traing_Data\images\CLIP071F00157.jpg 3099 | ..\Traing_Data\images\CLIP075F00048.jpg 3100 | ..\Traing_Data\images\CLIP075F00482.jpg 3101 | ..\Traing_Data\images\2600.jpg 3102 | ..\Traing_Data\images\1507.jpg 3103 | ..\Traing_Data\images\CLIP073F00068.jpg 3104 | ..\Traing_Data\images\4117.jpg 3105 | ..\Traing_Data\images\CLIP010729.jpg 3106 | ..\Traing_Data\images\CLIP012842.jpg 3107 | ..\Traing_Data\images\CLIP072F00122.jpg 3108 | ..\Traing_Data\images\CLIP077F00299.jpg 3109 | ..\Traing_Data\images\CLIP002726.jpg 3110 | ..\Traing_Data\images\CLIP064F00096.jpg 3111 | ..\Traing_Data\images\457.jpg 3112 | ..\Traing_Data\images\1182.jpg 3113 | ..\Traing_Data\images\CLIP077F01562.jpg 3114 | ..\Traing_Data\images\1773.jpg 3115 | ..\Traing_Data\images\CLIP068F00103.jpg 3116 | ..\Traing_Data\images\CLIP072F00924.jpg 3117 | ..\Traing_Data\images\CLIP075F00047.jpg 3118 | ..\Traing_Data\images\584.jpg 3119 | ..\Traing_Data\images\CLIP074F00364.jpg 3120 | ..\Traing_Data\images\CLIP072F00151.jpg 3121 | ..\Traing_Data\images\649.jpg 3122 | ..\Traing_Data\images\CLIP075F00422.jpg 3123 | ..\Traing_Data\images\CLIP072F00087.jpg 3124 | ..\Traing_Data\images\CLIP073F00804.jpg 3125 | ..\Traing_Data\images\CLIP074F00297.jpg 3126 | ..\Traing_Data\images\3643.jpg 3127 | ..\Traing_Data\images\CLIP008196.jpg 3128 | ..\Traing_Data\images\CLIP013848.jpg 3129 | ..\Traing_Data\images\CLIP010075.jpg 3130 | ..\Traing_Data\images\CLIP077F00500.jpg 3131 | ..\Traing_Data\images\2442.jpg 3132 | ..\Traing_Data\images\CLIP013247.jpg 3133 | ..\Traing_Data\images\CLIP077F00816.jpg 3134 | ..\Traing_Data\images\CLIP012839.jpg 3135 | ..\Traing_Data\images\950.jpg 3136 | ..\Traing_Data\images\CLIP072F00099.jpg 3137 | ..\Traing_Data\images\CLIP002373.jpg 3138 | ..\Traing_Data\images\CLIP075F00758.jpg 3139 | ..\Traing_Data\images\CLIP005143.jpg 3140 | ..\Traing_Data\images\CLIP074F00296.jpg 3141 | ..\Traing_Data\images\CLIP075F00748.jpg 3142 | ..\Traing_Data\images\CLIP073F00059.jpg 3143 | ..\Traing_Data\images\1827.jpg 3144 | ..\Traing_Data\images\CLIP075F00509.jpg 3145 | ..\Traing_Data\images\CLIP003927.jpg 3146 | ..\Traing_Data\images\CLIP005688.jpg 3147 | ..\Traing_Data\images\318.jpg 3148 | ..\Traing_Data\images\CLIP012278.jpg 3149 | ..\Traing_Data\images\CLIP073F00897.jpg 3150 | ..\Traing_Data\images\CLIP071F00188.jpg 3151 | ..\Traing_Data\images\CLIP070F00036.jpg 3152 | ..\Traing_Data\images\CLIP008751.jpg 3153 | ..\Traing_Data\images\CLIP071F00150.jpg 3154 | ..\Traing_Data\images\CLIP008232.jpg 3155 | ..\Traing_Data\images\CLIP000736.jpg 3156 | ..\Traing_Data\images\CLIP011726.jpg 3157 | ..\Traing_Data\images\CLIP071F00291.jpg 3158 | ..\Traing_Data\images\CLIP000100.jpg 3159 | ..\Traing_Data\images\CLIP072F00163.jpg 3160 | ..\Traing_Data\images\CLIP010646.jpg 3161 | ..\Traing_Data\images\CLIP077F00813.jpg 3162 | ..\Traing_Data\images\CLIP072F00140.jpg 3163 | ..\Traing_Data\images\CLIP006534.jpg 3164 | ..\Traing_Data\images\CLIP012057.jpg 3165 | ..\Traing_Data\images\CLIP000702.jpg 3166 | ..\Traing_Data\images\CLIP073F00654.jpg 3167 | ..\Traing_Data\images\CLIP078F00972.jpg 3168 | ..\Traing_Data\images\CLIP004538.jpg 3169 | ..\Traing_Data\images\CLIP003756.jpg 3170 | ..\Traing_Data\images\CLIP069F00110.jpg 3171 | ..\Traing_Data\images\CLIP072F00040.jpg 3172 | ..\Traing_Data\images\2913.jpg 3173 | ..\Traing_Data\images\CLIP002980.jpg 3174 | ..\Traing_Data\images\CLIP076F00536.jpg 3175 | ..\Traing_Data\images\CLIP069F00099.jpg 3176 | ..\Traing_Data\images\3128.jpg 3177 | ..\Traing_Data\images\CLIP070F00021.jpg 3178 | ..\Traing_Data\images\CLIP002061.jpg 3179 | ..\Traing_Data\images\CLIP074F00620.jpg 3180 | ..\Traing_Data\images\1786.jpg 3181 | ..\Traing_Data\images\CLIP001761.jpg 3182 | ..\Traing_Data\images\CLIP013067.jpg 3183 | ..\Traing_Data\images\2526.jpg 3184 | ..\Traing_Data\images\2123.jpg 3185 | ..\Traing_Data\images\3050.jpg 3186 | ..\Traing_Data\images\CLIP012829.jpg 3187 | ..\Traing_Data\images\728.jpg 3188 | ..\Traing_Data\images\CLIP001259.jpg 3189 | ..\Traing_Data\images\CLIP072F00956.jpg 3190 | ..\Traing_Data\images\CLIP075F00449.jpg 3191 | ..\Traing_Data\images\CLIP072F00036.jpg 3192 | ..\Traing_Data\images\CLIP011819.jpg 3193 | ..\Traing_Data\images\CLIP007370.jpg 3194 | ..\Traing_Data\images\2527.jpg 3195 | ..\Traing_Data\images\CLIP064F00355.jpg 3196 | ..\Traing_Data\images\CLIP070F00074.jpg 3197 | ..\Traing_Data\images\CLIP075F00459.jpg 3198 | ..\Traing_Data\images\CLIP072F00699.jpg 3199 | ..\Traing_Data\images\CLIP074F00353.jpg 3200 | ..\Traing_Data\images\785.jpg 3201 | ..\Traing_Data\images\CLIP064F00138.jpg 3202 | ..\Traing_Data\images\CLIP069F00177.jpg 3203 | ..\Traing_Data\images\CLIP079F00454.jpg 3204 | ..\Traing_Data\images\CLIP011910.jpg 3205 | ..\Traing_Data\images\CLIP072F00917.jpg 3206 | ..\Traing_Data\images\CLIP075F00370.jpg 3207 | ..\Traing_Data\images\1585.jpg 3208 | ..\Traing_Data\images\CLIP007557.jpg 3209 | ..\Traing_Data\images\CLIP006133.jpg 3210 | ..\Traing_Data\images\CLIP012642.jpg 3211 | ..\Traing_Data\images\CLIP010246.jpg 3212 | ..\Traing_Data\images\CLIP064F00399.jpg 3213 | ..\Traing_Data\images\CLIP073F00972.jpg 3214 | ..\Traing_Data\images\CLIP012652.jpg 3215 | ..\Traing_Data\images\587.jpg 3216 | ..\Traing_Data\images\CLIP079F00070.jpg 3217 | ..\Traing_Data\images\CLIP003060.jpg 3218 | ..\Traing_Data\images\CLIP008869.jpg 3219 | ..\Traing_Data\images\CLIP070F00063.jpg 3220 | ..\Traing_Data\images\1697.jpg 3221 | ..\Traing_Data\images\CLIP012881.jpg 3222 | ..\Traing_Data\images\CLIP011224.jpg 3223 | ..\Traing_Data\images\CLIP007973.jpg 3224 | ..\Traing_Data\images\CLIP067F00022.jpg 3225 | ..\Traing_Data\images\CLIP073F00002.jpg 3226 | ..\Traing_Data\images\CLIP075F00528.jpg 3227 | ..\Traing_Data\images\CLIP012894.jpg 3228 | ..\Traing_Data\images\1030.jpg 3229 | ..\Traing_Data\images\CLIP072F00499.jpg 3230 | ..\Traing_Data\images\CLIP077F00477.jpg 3231 | ..\Traing_Data\images\CLIP004602.jpg 3232 | ..\Traing_Data\images\CLIP069F00172.jpg 3233 | ..\Traing_Data\images\CLIP001111.jpg 3234 | ..\Traing_Data\images\CLIP073F00422.jpg 3235 | ..\Traing_Data\images\CLIP005426.jpg 3236 | ..\Traing_Data\images\CLIP064F00125.jpg 3237 | ..\Traing_Data\images\CLIP073F00001.jpg 3238 | ..\Traing_Data\images\CLIP072F00100.jpg 3239 | ..\Traing_Data\images\CLIP006125.jpg 3240 | ..\Traing_Data\images\269.jpg 3241 | ..\Traing_Data\images\CLIP009605.jpg 3242 | ..\Traing_Data\images\CLIP013121.jpg 3243 | ..\Traing_Data\images\CLIP004297.jpg 3244 | ..\Traing_Data\images\CLIP012220.jpg 3245 | ..\Traing_Data\images\CLIP076F00329.jpg 3246 | ..\Traing_Data\images\CLIP072F00548.jpg 3247 | ..\Traing_Data\images\CLIP003457.jpg 3248 | ..\Traing_Data\images\CLIP073F00887.jpg 3249 | ..\Traing_Data\images\CLIP074F00467.jpg 3250 | ..\Traing_Data\images\CLIP003529.jpg 3251 | ..\Traing_Data\images\CLIP072F00513.jpg 3252 | ..\Traing_Data\images\CLIP077F01559.jpg 3253 | ..\Traing_Data\images\CLIP074F00169.jpg 3254 | ..\Traing_Data\images\CLIP004533.jpg 3255 | ..\Traing_Data\images\CLIP071F00226.jpg 3256 | ..\Traing_Data\images\CLIP068F00039.jpg 3257 | ..\Traing_Data\images\CLIP071F00315.jpg 3258 | ..\Traing_Data\images\CLIP012538.jpg 3259 | ..\Traing_Data\images\CLIP013304.jpg 3260 | ..\Traing_Data\images\CLIP005776.jpg 3261 | ..\Traing_Data\images\CLIP071F00287.jpg 3262 | ..\Traing_Data\images\CLIP003749.jpg 3263 | ..\Traing_Data\images\2327.jpg 3264 | ..\Traing_Data\images\CLIP079F00809.jpg 3265 | --------------------------------------------------------------------------------