CVPR 2024 accepted papers on Video

271 of the 2,716 papers accepted at CVPR 2024 (10.0%) are on Video, compared with 10.1% at CVPR 2023. They fall into 4 topics.

See CVPR 2024 in the interactive map

Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.

Topics and papers

Video segmentation and tracking · 110 papers

Neural Video Compression with Feature Modulation (140 citations)
Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (104 citations)
Open-Vocabulary Video Anomaly Detection (90 citations)
Exploiting Style Latent Flows for Generalizing Deepfake Video Detection (77 citations)
Self-Distilled Masked Auto-Encoders are Efficient Video Anomaly Detectors (64 citations)
MemSAM: Taming Segment Anything Model for Echocardiography Video Segmentation (62 citations)
C3: High-Performance and Low-Complexity Neural Compression from a Single Image or Video (50 citations)
CoDeF: Content Deformation Fields for Temporally Consistent Video Processing (48 citations)
Generalized Predictive Model for Autonomous Driving (47 citations)
MemFlow: Optical Flow Estimation and Prediction with Memory (45 citations)
Prompt-Enhanced Multiple Instance Learning for Weakly Supervised Video Anomaly Detection (44 citations)
Decoupling Static and Hierarchical Motion Perception for Referring Video Segmentation (41 citations)

and 98 more in the interactive map.

Video generation · 66 papers

VBench: Comprehensive Benchmark Suite for Video Generative Models (351 citations)
VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models (223 citations)
MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model (143 citations)
Video-P2P: Video Editing with Cross-attention Control (123 citations)
Animate Anyone: Consistent and Controllable Image-to-Video Synthesis for Character Animation (102 citations)
EvalCrafter: Benchmarking and Evaluating Large Video Generation Models (90 citations)
DisCo: Disentangled Control for Realistic Human Dance Generation (79 citations)
Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution (78 citations)
MMM: Generative Masked Motion Model (63 citations)
Lodge: A Coarse to Fine Diffusion Network for Long Dance Generation Guided by the Characteristic Dance Primitives (56 citations)
Generative Image Dynamics (48 citations)
SimDA: Simple Diffusion Adapter for Efficient Video Generation (45 citations)

and 54 more in the interactive map.

Video-language understanding · 62 papers

MVBench: A Comprehensive Multi-modal Video Understanding Benchmark (241 citations)
MovieChat: From Dense Token to Sparse Memory for Long Video Understanding (154 citations)
TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding (130 citations)
VTimeLLM: Empower LLM to Grasp Video Moments (119 citations)
Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives (105 citations)
Text Prompt with Normality Guidance for Weakly Supervised Video Anomaly Detection (95 citations)
Harnessing Large Language Models for Training-free Video Anomaly Detection (89 citations)
Panda-70M: Captioning 70M Videos with Multiple Cross-Modality Teachers (81 citations)
MA-LMM: Memory-Augmented Large Multimodal Model for Long-Term Video Understanding (80 citations)
Text Is MASS: Modeling as Stochastic Embedding for Text-Video Retrieval (77 citations)
VideoLLM-online: Online Video Large Language Model for Streaming Video (65 citations)
Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection (61 citations)

and 50 more in the interactive map.

Action recognition · 33 papers

and 21 more in the interactive map.