CVPR 2022 accepted papers on Video
222 of the 2,077 papers accepted at CVPR 2022 (10.7%) are on Video, compared with 10.2% at CVPR 2021. They fall into 4 topics.
See CVPR 2022 in the interactive map
Submitting to CVPR? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Video segmentation and tracking · 120 papers
- BasicVSR++: Improving Video Super-Resolution With Enhanced Propagation and Alignment (586 citations)
- Masked Feature Prediction for Self-Supervised Visual Pre-Training (559 citations)
- SimVP: Simpler Yet Better Video Prediction (393 citations)
- PhysFormer: Facial Video-Based Physiological Measurement With Temporal Difference Transformer (293 citations)
- Clothes-Changing Person Re-Identification With RGB Modality Only (253 citations)
- Generative Cooperative Learning for Unsupervised Video Anomaly Detection (226 citations)
- BEVT: BERT Pretraining of Video Transformers (214 citations)
- IFRNet: Intermediate Feature Refine Network for Efficient Frame Interpolation (193 citations)
- StyleGAN-V: A Continuous Video Generator With the Price, Image Quality and Perks of StyleGAN2 (190 citations)
- UBnormal: New Benchmark for Supervised Open-Set Video Anomaly Detection (183 citations)
- Leveraging Real Talking Faces via Self-Supervision for Robust Forgery Detection (166 citations)
- Towards an End-to-End Framework for Flow-Guided Video Inpainting (145 citations)
and 108 more in the interactive map.
Action recognition · 65 papers
- Video Swin Transformer (2,080 citations)
- Multiview Transformers for Video Recognition (292 citations)
- Assembly101: A Large-Scale Multi-View Video Dataset for Understanding Procedural Activities (203 citations)
- MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video Recognition (171 citations)
- Spatio-Temporal Relation Modeling for Few-Shot Action Recognition (159 citations)
- FineDiving: A Fine-Grained Dataset for Procedure-Aware Action Quality Assessment (143 citations)
- Hybrid Relation Guided Set Matching for Few-Shot Action Recognition (138 citations)
- Recurring the Transformer for Video Action Recognition (131 citations)
- Animal Kingdom: A Large and Diverse Dataset for Animal Behavior Understanding (125 citations)
- Self-Supervised Video Transformer (119 citations)
- ASM-Loc: Action-Aware Segment Modeling for Weakly-Supervised Temporal Action Localization (114 citations)
- MS-TCT: Multi-Scale Temporal ConvTransformer for Action Detection (108 citations)
and 53 more in the interactive map.
Video-language understanding · 29 papers
- SwinBERT: End-to-End Transformers With Sparse Attention for Video Captioning (296 citations)
- X-Pool: Cross-Modal Language-Video Attention for Text-Video Retrieval (223 citations)
- UMT: Unified Multi-Modal Transformers for Joint Video Moment Retrieval and Highlight Detection (173 citations)
- End-to-End Referring Video Object Segmentation With Multimodal Transformers (167 citations)
- End-to-End Generative Pretraining for Multimodal Video Captioning (158 citations)
- Revisiting the "Video" in Video-Language Understanding (134 citations)
- Bridging Video-Text Retrieval With Multiple Choice Questions (129 citations)
- Weakly Supervised Temporal Sentence Grounding With Gaussian-Based Contrastive Proposal Learning (120 citations)
- Advancing High-Resolution Video-Language Representation With Large-Scale Video Transcriptions (117 citations)
- Hierarchical Modular Network for Video Captioning (102 citations)
- TubeDETR: Spatio-Temporal Video Grounding With Transformers (94 citations)
- MAD: A Scalable Dataset for Language Grounding in Videos From Movie Audio Descriptions (91 citations)
and 17 more in the interactive map.
Video generation · 8 papers
- FaceFormer: Speech-Driven 3D Facial Animation With Transformers (252 citations)
- Make It Move: Controllable Image-to-Video Generation With Text Descriptions (60 citations)
- Show Me What and Tell Me How: Video Synthesis via Multimodal Conditioning (37 citations)
- Audio-Driven Neural Gesture Reenactment With Video Motion Graphs (25 citations)
- Speech Driven Tongue Animation (14 citations)
- More Than Words: In-the-Wild Visually-Driven Prosody for Text-to-Speech (13 citations)
- Playable Environments: Video Manipulation in Space and Time (13 citations)
- TemporalUV: Capturing Loose Clothing With Temporally Coherent UV Coordinates (6 citations)
