└── README.md /README.md: -------------------------------------------------------------------------------- 1 | # Awesome Causality in Computer Vision [![Awesome](https://awesome.re/badge.svg)](https://awesome.re) 2 | 3 | A curated list of **causality in computer vision**. Inspired by [awesome-deep-vision](https://github.com/kjw0612/awesome-deep-vision), [awesome-adversarial-machine-learning](https://github.com/yenchenlin/awesome-adversarial-machine-learning), [awesome-deep-learning-papers](https://github.com/terryum/awesome-deep-learning-papers) and [Awesome-NAS](https://github.com/D-X-Y/Awesome-NAS). 4 | 5 | Please feel free to [pull requests](https://github.com/he-y/awesome-Pruning/pulls) or [open an issue](https://github.com/he-y/awesome-Pruning/issues) to add papers. 6 | 7 | ## Table of Contents 8 | 9 | - [Type of Causality in CV](#type-of-causality-in-cv) 10 | - [2021 Venues](#2021) 11 | - [2020 Venues](#2020) 12 | - [Before 2020](#before-2020) 13 | - [Arxiv](#arxiv) 14 | 15 | 16 | ### Type of Causality in CV 17 | 18 | | Type | `IT` | `CF` | `CR` | `O` | 19 | |:----------- |:--------------:|:--------------:|:-----------:|:-----------:| 20 | | Explanation | Intervention | Counterfactual | Causal Representation | Other Types | 21 | 22 | 23 | ### 2021 24 | 25 | | Title | Venue | Type | Code | 26 | |:-------------------------------------------------------------------------------------------------------------------------------- |:-----:|:-------:|:----:| 27 | | [Transporting Causal Mechanisms for Unsupervised Domain Adaptation](https://arxiv.org/abs/2107.11055) | ICCV | `O` | [PyTorch(Author)](https://github.com/yue-zhongqi/tcm) | 28 | | [Causal Attention for Unbiased Visual Recognition](https://arxiv.org/abs/2108.08782) | ICCV | `IT` | [PyTorch(Author)](https://github.com/Wangt-CN/CaaM) | 29 | | [Learning Causal Representation for Training Cross-Domain Pose Estimator via Generative Interventions](https://openaccess.thecvf.com/content/ICCV2021/papers/Zhang_Learning_Causal_Representation_for_Training_Cross-Domain_Pose_Estimator_via_Generative_ICCV_2021_paper.pdf) | ICCV | `IT/CR` | - | 30 | | [Human Trajectory Prediction via Counterfactual Analysis](https://arxiv.org/abs/2107.14202) | ICCV | `CF` | [PyTorch(Author)](https://github.com/CHENGY12/CausalHTP) | 31 | | [Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-Identification](https://arxiv.org/abs/2108.08728) | ICCV | `CF` | [PyTorch(Author)](https://github.com/raoyongming/cal) | 32 | | [Counterfactual VQA: A Cause-Effect Look at Language Bias](https://arxiv.org/abs/2006.04315) | CVPR | `CF` | [PyTorch(Author)](https://github.com/yuleiniu/cfvqa) | 33 | | [Counterfactual Zero-Shot and Open-Set Visual Recognition](https://arxiv.org/abs/2103.00887) | CVPR | `CF` | [PyTorch(Author)](https://github.com/yue-zhongqi/gcm-cf) | 34 | | [Distilling Causal Effect of Data in Class-Incremental Learning](https://arxiv.org/pdf/2103.01737) | CVPR | `CF` | [PyTorch(Author)](https://github.com/JoyHuYY1412/DDE_CIL) | 35 | | [Causal Attention for Vision-Language Tasks](https://arxiv.org/abs/2103.03493) | CVPR | `IT` | [PyTorch(Author)](https://github.com/yangxuntu/lxmertcatt) | 36 | | [The Blessings of Unlabeled Background in Untrimmed Videos](https://arxiv.org/abs/2103.13183) | CVPR | `CF` | - | 37 | | [Affect2MM: Affective Analysis of Multimedia Content Using Emotion Causality](https://arxiv.org/abs/2103.06541) | CVPR | `O` | [PyTorch(Author)](https://gamma.umd.edu/researchdirections/affectivecomputing/affect2mm/)| 38 | | [CausalVAE: Disentangled Representation Learning via Neural Structural Causal Models](https://arxiv.org/abs/2004.08697) | CVPR | `CR` | - | 39 | | [Generative Interventions for Causal Learning](https://arxiv.org/abs/2012.12265) | CVPR | `CR` | [PyTorch(Author)](https://github.com/cvlab-columbia/GenInt) | 40 | | [ACRE: Abstract Causal REasoning Beyond Covariation](https://arxiv.org/abs/2103.14232) | CVPR | `O` | [PyTorch(Author)](https://github.com/WellyZhang/ACRE) | 41 | | [Towards Robust Classification Model by Counterfactual and Invariant Data Generation](https://arxiv.org/abs/2106.01127) | CVPR | `CF` | [PyTorch(Author)](https://github.com/WellyZhang/ACRE) | 42 | | [Interventional Video Grounding With Dual Contrastive Learning](https://arxiv.org/abs/2106.11013) | CVPR | `IT` | - | 43 | | [Representation Learning via Invariant Causal Mechanisms](https://arxiv.org/abs/2010.07922) | ICLR | `CR` | - | 44 | | [Counterfactual Generative Networks](https://arxiv.org/abs/2101.06046) | ICLR | `CF` | [PyTorch(Author)](https://github.com/autonomousvision/counterfactual_generative_networks) | 45 | 46 | 47 | 48 | ### 2020 49 | 50 | | Title | Venue | Type | Code | 51 | |:-------------------------------------------------------------------------------------------------------------------------------- |:-----:|:-------:|:----:| 52 | | [Counterfactual Contrastive Learning for Weakly-Supervised Vision-Language Grounding](https://papers.nips.cc/paper/2020/file/d27b95cac4c27feb850aaa4070cc4675-Paper.pdf) | NeurIPS | `CF` | - | 53 | | [A causal view of compositional zero-shot recognition](https://arxiv.org/abs/2006.14610) | NeurIPS | `IT` | - | 54 | | [Counterfactual Vision-and-Language Navigation: Unravelling the Unseen](https://papers.nips.cc/paper/2020/hash/39016cfe079db1bfb359ca72fcba3fd8-Abstract.html) | NeurIPS | `CF` | - | 55 | | [Causal Intervention for Weakly-Supervised Semantic Segmentation](https://arxiv.org/abs/2009.12547) | NeurIPS | `IT` | [PyTorch(Author)](https://github.com/ZHANGDONG-NJUST/CONTA) | 56 | | [Interventional Few-Shot Learning](http://arxiv.org/abs/2009.13000) | NeurIPS | `IT` | [PyTorch(Author)](https://github.com/yue-zhongqi/ifsl) | 57 | | [Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effect](https://arxiv.org/abs/2009.12991) | NeurIPS | `CF` | [PyTorch(Author)](https://github.com/KaihuaTang/Long-Tailed-Recognition.pytorch) | 58 | | [Traffic Accident Benchmark for Causality Recognition](https://arxiv.org/abs/1911.07308) | ECCV | `O` | [PyTorch(Author)](https://github.com/tackgeun/CausalityInTrafficAccident) | 59 | | [Towards causal benchmarking of bias in face analysis algorithms](https://arxiv.org/abs/2007.06570) | ECCV | `O` | [PyTorch(Author)](https://github.com/balakg/transects-eccv2020) | 60 | | [Learning What Makes a Difference from Counterfactual Examples and Gradient Supervision](https://arxiv.org/abs/2004.09034) | ECCV | `CF` | - | 61 | | [Counterfactual Vision-and-Language Navigation via Adversarial Path Sampling](https://arxiv.org/abs/1911.07308) | ECCV | `CF` | - | 62 | | [Visual Commonsense R-CNN](https://arxiv.org/abs/2002.12204) | CVPR | `CR/IT` | [PyTorch(Author)](https://github.com/Wangt-CN/VC-R-CNN) | 63 | | [Unbiased scene graph generation from biased training](https://arxiv.org/abs/2002.11949) | CVPR | `CF` | [PyTorch(Author)](https://github.com/KaihuaTang/Scene-Graph-Benchmark.pytorch) | 64 | | [Two causal principles for improving visual dialog](https://arxiv.org/abs/1911.10496) | CVPR | `IT` | [PyTorch(Author)](https://github.com/simpleshinobu/visdial-principles) | 65 | | [Counterfactual samples synthesizing for robust visual question answering](https://openaccess.thecvf.com/content_CVPR_2020/papers/Chen_Counterfactual_Samples_Synthesizing_for_Robust_Visual_Question_Answering_CVPR_2020_paper.pdf) | CVPR | `CF` | [PyTorch(Author)](https://github.com/yanxinzju/CSS-VQA) | 66 | | [Towards Causal VQA: Revealing and Reducing Spurious Correlations by Invariant and Covariant Semantic Editing](https://arxiv.org/abs/1912.07538) | CVPR | `CF` | [PyTorch(Author)](https://github.com/AgarwalVedika/CausalVQA) | 67 | | [Counterfactuals uncover the modular structure of deep generative models](https://arxiv.org/abs/1912.07538) | ICLR | `CF` | - | 68 | | [Exploratory Not Explanatory: Counterfactual Analysis of Saliency Maps for Deep Reinforcement Learning](https://arxiv.org/abs/1912.05743) | ICLR | `CF` | - | 69 | 70 | 71 | 72 | ### Before 2020 73 | 74 | | Title | Venue | Type | Code | 75 | |:-------|:--------:|:-------:|:-------:| 76 | | [Counterfactual Visual Explanations](https://arxiv.org/abs/1904.07451) | ICML2019 | `CF` | - | 77 | | [CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training](https://arxiv.org/abs/1912.07538) |ICLR2018 | `O` | [PyTorch(Author)](https://github.com/mkocaoglu/CausalGAN) | 78 | | [Discovering causal signals in images](https://openaccess.thecvf.com/content_cvpr_2017/papers/Lopez-Paz_Discovering_Causal_Signals_CVPR_2017_paper.pdf) | CVPR2017 | `O` | [TensorFlow(3rd)](https://github.com/kyrs/NCC-experiments) | 79 | 80 | 81 | 82 | 83 | ### Arxiv 84 | 85 | | Title | Venue | Type | Code | 86 | |:-------------------------------------------------------------------------------------------------------------------------------- |:-----:|:-------:|:----:| 87 | | [Learning Causal Semantic Representation for Out-of-Distribution Prediction](https://arxiv.org/abs/2011.01681) | Arxiv | `CR` | - | 88 | | [ECINN: Efficient Counterfactuals from Invertible Neural Networks](https://arxiv.org/abs/2103.13701) | Arxiv | `CF` | - | 89 | | [A Structural Causal Model for MR Images of Multiple Sclerosis](https://arxiv.org/abs/2103.03158) | Arxiv | `CF` | - | 90 | | [Counterfactual Variable Control for Robust and Interpretable Question Answering](https://arxiv.org/abs/2010.05581) | Arxiv | `CF` | [PyTorch(Author)](https://github.com/PluviophileYU/CVC-QA) | 91 | | [Deconfounded Image Captioning: A Causal Retrospect](https://arxiv.org/abs/2003.03923) | Arxiv | `IT` | - | 92 | | [Latent Causal Invariant Model](https://arxiv.org/abs/2011.02203) | Arxiv | `O` | - | 93 | 94 | 95 | 96 | 97 |
98 | 99 | ## Related Repo 100 | 101 | [awesome-causality-algorithms](https://github.com/rguo12/awesome-causality-algorithms) 102 | 103 | [Causal_Reading_Group](https://github.com/fulifeng/Causal_Reading_Group) 104 | 105 | [awesome-causality](https://github.com/napsternxg/awesome-causality) 106 | 107 | 108 | --------------------------------------------------------------------------------