ICCV 2023 accepted papers on Video
249 of the 2,156 papers accepted at ICCV 2023 (11.5%) are on Video, compared with 11.6% at ICCV 2021. They fall into 4 topics.
See ICCV 2023 in the interactive map
Submitting to ICCV? 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 · 104 papers
- Tracking Anything with Decoupled Video Segmentation (228 citations)
- MOSE: A New Dataset for Video Object Segmentation in Complex Scenes (176 citations)
- Tracking Everything Everywhere All at Once (174 citations)
- ProPainter: Improving Propagation and Transformer for Video Inpainting (149 citations)
- TALL: Thumbnail Layout for Deepfake Video Detection (134 citations)
- MeViS: A Large-scale Benchmark for Video Segmentation with Motion Expressions (133 citations)
- VideoFlow: Exploiting Temporal Cues for Multi-frame Optical Flow Estimation (128 citations)
- Feature Prediction Diffusion Model for Video Anomaly Detection (102 citations)
- Multimodal Motion Conditioned Diffusion Model for Skeleton-based Video Anomaly Detection (98 citations)
- UniFormerV2: Unlocking the Potential of Image ViTs for Video Understanding (94 citations)
- MMST-ViT: Climate Change-aware Crop Yield Prediction via Multi-Modal Spatial-Temporal Vision Transformer (86 citations)
- DVIS: Decoupled Video Instance Segmentation Framework (81 citations)
and 92 more in the interactive map.
Video generation · 53 papers
- Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation (801 citations)
- Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generators (625 citations)
- Structure and Content-Guided Video Synthesis with Diffusion Models (526 citations)
- FateZero: Fusing Attentions for Zero-shot Text-based Video Editing (390 citations)
- PhysDiff: Physics-Guided Human Motion Diffusion Model (313 citations)
- Exploring Video Quality Assessment on User Generated Contents from Aesthetic and Technical Perspectives (304 citations)
- ReMoDiffuse: Retrieval-Augmented Motion Diffusion Model (241 citations)
- Pix2Video: Video Editing using Image Diffusion (235 citations)
- Guided Motion Diffusion for Controllable Human Motion Synthesis (221 citations)
- Preserve Your Own Correlation: A Noise Prior for Video Diffusion Models (176 citations)
- TMR: Text-to-Motion Retrieval Using Contrastive 3D Human Motion Synthesis (150 citations)
- StableVideo: Text-driven Consistency-aware Diffusion Video Editing (145 citations)
and 41 more in the interactive map.
Action recognition · 48 papers
- Generative Action Description Prompts for Skeleton-based Action Recognition (142 citations)
- Diffusion Action Segmentation (134 citations)
- DyGait: Exploiting Dynamic Representations for High-performance Gait Recognition (85 citations)
- Masked Motion Predictors are Strong 3D Action Representation Learners (76 citations)
- Memory-and-Anticipation Transformer for Online Action Understanding (67 citations)
- SKiT: a Fast Key Information Video Transformer for Online Surgical Phase Recognition (63 citations)
- How Much Temporal Long-Term Context is Needed for Action Segmentation? (58 citations)
- Cross-Modal Learning with 3D Deformable Attention for Action Recognition (56 citations)
- Action Sensitivity Learning for Temporal Action Localization (53 citations)
- Hierarchical Spatio-Temporal Representation Learning for Gait Recognition (48 citations)
- SkeleTR: Towards Skeleton-based Action Recognition in the Wild (48 citations)
- Boosting Few-shot Action Recognition with Graph-guided Hybrid Matching (47 citations)
and 36 more in the interactive map.
Video-language understanding · 44 papers
- UniVTG: Towards Unified Video-Language Temporal Grounding (198 citations)
- Unmasked Teacher: Towards Training-Efficient Video Foundation Models (193 citations)
- EgoVLPv2: Egocentric Video-Language Pre-training with Fusion in the Backbone (107 citations)
- UATVR: Uncertainty-Adaptive Text-Video Retrieval (95 citations)
- Unified Coarse-to-Fine Alignment for Video-Text Retrieval (95 citations)
- Knowing Where to Focus: Event-aware Transformer for Video Grounding (91 citations)
- HiTeA: Hierarchical Temporal-Aware Video-Language Pre-training (89 citations)
- Verbs in Action: Improving Verb Understanding in Video-Language Models (73 citations)
- Progressive Spatio-Temporal Prototype Matching for Text-Video Retrieval (70 citations)
- UnLoc: A Unified Framework for Video Localization Tasks (63 citations)
- Prompt Switch: Efficient CLIP Adaptation for Text-Video Retrieval (56 citations)
- Audio-Enhanced Text-to-Video Retrieval using Text-Conditioned Feature Alignment (43 citations)
and 32 more in the interactive map.
