└── README.md


/README.md:
--------------------------------------------------------------------------------
 1 | # rvqa
 2 | 
 3 | The Relation-VQA dataset is released!
 4 | 
 5 | If you have any question, please do not hesitate to contact me.
 6 | 
 7 | 
 8 | ## Relation-VQA Dataset
 9 | The dataset can be downloaded from the links: [[Link 1](https://drive.google.com/file/d/15zrDnagK3qgKYGkVUrnwr0Zrf2Y8VlSA)], [[Link 2](https://drive.google.com/file/d/1x2T7R6XB3pmqCstffhK_Zv7CkKui95wB)]. 
10 | 
11 | Or you can download the data files one by one:
12 | 
13 | **The whole dataset**
14 | - [vg_vqa-rel_map_result_0.30.json](https://drive.google.com/file/d/1YoYMvsg7WweMrz1zC-hPFcTNi21G_Srv)
15 | - [vg_vqa-rel_map_result_0.30.txt](https://drive.google.com/file/d/1RJIFFp20x5lXbpGgWxb3AXa3I8ONfoJ4)
16 | 
17 | **The splitted dataset**
18 | - [vqa_raw_train_0.30.json](https://drive.google.com/file/d/1R3Hdkqx2sSw2B8Z4VIWQzkrjLfJ7-76y)
19 | - [vqa_raw_val_0.30.json](https://drive.google.com/file/d/18K9WjLu4hN1gEgrJaPSlvZopORlxsGtk)
20 | - [vqa_raw_train_val_0.30.json](https://drive.google.com/file/d/1YDhjFxUzd6zRFJxXA7V968eE0PUk8f-l)
21 | - [vqa_raw_test_0.30.json](https://drive.google.com/file/d/1HPjR6uSaUVeftVKwuejRkb_SnmNvm7RK)
22 | 
23 | **Alias rules**
24 | - [alias_map_dict.json](https://drive.google.com/file/d/1tnEZPtu8lwjFM0WwmpfTE3DIyb_eSF8z)
25 | 
26 | ### Reference
27 | - **Paper on SIGKDD**: http://www.kdd.org/kdd2018/accepted-papers/view/r-vqa-learning-visual-relation-facts-with-semantic-attention-for-visual-que
28 | - **Paper on arXiv**: https://arxiv.org/abs/1805.09701
29 | 
30 | 
31 | 
32 | If you use this project as part of any published research, please acknowledge the following paper.
33 | ```
34 | @inproceedings{lu2018rvqa,
35 | 	title={R-VQA: Learning Visual Relation Facts with Semantic Attention for Visual Question Answering.},
36 | 	author={Lu, Pan and Ji, Lei and Zhang, Wei and Duan, Nan and Zhou, Ming and Wang, Jianyong},
37 | 	booktitle={SIGKDD 2018},
38 | 	year={2018}
39 | }
40 | ```
41 | 


--------------------------------------------------------------------------------