AAAI 2025 accepted papers on Video
175 of the 3,486 papers accepted at AAAI 2025 (5.0%) are on Video, compared with 3.9% at AAAI 2024. They fall into 4 topics.
See AAAI 2025 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 · 67 papers
- Decomposed Spatio-Temporal Mamba for Long-Term Traffic Prediction (24 citations)
- CLIMB-ReID: A Hybrid CLIP-Mamba Framework for Person Re-Identification (21 citations)
- Exploring More from Multiple Gait Modalities for Human Identification (21 citations)
- Bridging Traffic State and Trajectory for Dynamic Road Network and Trajectory Representation Learning (17 citations)
- STD-PLM: Understanding Both Spatial and Temporal Properties of Spatial-Temporal Data with PLM (17 citations)
- Video Anomaly Detection with Motion and Appearance Guided Patch Diffusion Model (16 citations)
- Depth-Centric Dehazing and Depth-Estimation from Real-World Hazy Driving Video (15 citations)
- DepMGNN: Matrixial Graph Neural Network for Video-based Automatic Depression Assessment (14 citations)
- MOCID: Motion Context and Displacement Information Learning for Moving Infrared Small Target Detection (14 citations)
- PNVC: Towards Practical INR-based Video Compression (14 citations)
- E4: Energy-Efficient DNN Inference for Edge Video Analytics via Early Exiting and DVFS (12 citations)
- DanceFix: An Exploration in Group Dance Neatness Assessment Through Fixing Abnormal Challenges of Human Pose (11 citations)
and 55 more in the interactive map.
Video-language understanding · 49 papers
- VTG-LLM: Integrating Timestamp Knowledge into Video LLMs for Enhanced Video Temporal Grounding (30 citations)
- V2Xum-LLM: Cross-Modal Video Summarization with Temporal Prompt Instruction Tuning (22 citations)
- Federated Weakly Supervised Video Anomaly Detection with Multimodal Prompt (16 citations)
- Dense Audio-Visual Event Localization Under Cross-Modal Consistency and Multi-Temporal Granularity Collaboration (13 citations)
- MUSE: Mamba Is Efficient Multi-scale Learner for Text-video Retrieval (9 citations)
- Empowering LLMs with Pseudo-Untrimmed Videos for Audio-Visual Temporal Understanding (8 citations)
- Text Proxy: Decomposing Retrieval from a 1-to-N Relationship into N 1-to-1 Relationships for Text-Video Retrieval (8 citations)
- Diversifying Query: Region-Guided Transformer for Temporal Sentence Grounding (7 citations)
- Bridging the Semantic Granularity Gap Between Text and Frame Representations for Partially Relevant Video Retrieval (6 citations)
- Generative Video Diffusion for Unseen Novel Semantic Video Moment Retrieval (6 citations)
- HiCM²: Hierarchical Compact Memory Modeling for Dense Video Captioning (6 citations)
- Ambiguity-Restrained Text-Video Representation Learning for Partially Relevant Video Retrieval (5 citations)
and 37 more in the interactive map.
Video generation · 45 papers
- DriveDreamer-2: LLM-Enhanced World Models for Diverse Driving Video Generation (59 citations)
- Sign-IDD: Iconicity Disentangled Diffusion for Sign Language Production (25 citations)
- VE-Bench: Subjective-Aligned Benchmark Suite for Text-Driven Video Editing Quality Assessment (21 citations)
- CustomCrafter: Customized Video Generation with Preserving Motion and Concept Composition Abilities (15 citations)
- Follow-Your-Click: Open-domain Regional Image Animation via Motion Prompts (15 citations)
- Read, Watch and Scream! Sound Generation from Text and Video (14 citations)
- Image Conductor: Precision Control for Interactive Video Synthesis (11 citations)
- CoCoCo: Improving Text-Guided Video Inpainting for Better Consistency, Controllability and Compatibility (10 citations)
- DreamPhysics: Learning Physics-Based 3D Dynamics with Video Diffusion Priors (10 citations)
- InstructAvatar: Text-Guided Emotion and Motion Control for Avatar Generation (9 citations)
- Video Diffusion Models Are Strong Video Inpainter (9 citations)
- ContextualStory: Consistent Visual Storytelling with Spatially-Enhanced and Storyline Context (8 citations)
and 33 more in the interactive map.
Action recognition · 14 papers
- Hierarchical Vector Quantization for Unsupervised Action Segmentation (13 citations)
- Frame Order Matters: A Temporal Sequence-Aware Model for Few-Shot Action Recognition (11 citations)
- Manta: Enhancing Mamba for Few-Shot Action Recognition of Long Sub-Sequence (9 citations)
- VA-AR: Learning Velocity-Aware Action Representations with Mixture of Window Attention (9 citations)
- Stable Mean Teacher for Semi-supervised Video Action Detection (7 citations)
- Prediction-Feedback DETR for Temporal Action Detection (6 citations)
- Surgical Workflow Recognition and Blocking Effectiveness Detection in Laparoscopic Liver Resection with Pringle Maneuver (6 citations)
- Action-Agnostic Point-Level Supervision for Temporal Action Detection (4 citations)
- Similar Modality Enhancement and Action Consistency Learning for Weakly Supervised Temporal Action Localization (4 citations)
- Temporal Action Localization with Cross Layer Task Decoupling and Refinement (4 citations)
- Building a Multi-modal Spatiotemporal Expert for Zero-shot Action Recognition with CLIP (3 citations)
- Stitch, Contrast, and Segment: Learning a Human Action Segmentation Model Using Trimmed Skeleton Videos (3 citations)
and 2 more in the interactive map.