AAAI 2026 accepted papers on Video
228 of the 4,417 papers accepted at AAAI 2026 (5.2%) are on Video, compared with 5.0% at AAAI 2025. They fall into 4 topics.
See AAAI 2026 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 generation · 79 papers
- VQ-Insight: Teaching VLMs for AI-Generated Video Quality Understanding via Progressive Visual Reinforcement Learning (5 citations)
- ContextFlow: Training-Free Video Object Editing via Adaptive Context Enrichment (4 citations)
- Mitigating Error Accumulation in Co-Speech Motion Generation via Global Rotation Diffusion and Multi-Level Constraints (4 citations)
- Pre-Trained Video Generative Models as World Simulators (4 citations)
- GENMAC: Compositional Text-to-Video Generation with Multi-Agent Collaboration (3 citations)
- ReAlign: Text-to-Motion Generation via Step-Aware Reward-Guided Alignment (3 citations)
- Sparse-vDiT: Unleashing the Power of Sparse Attention to Accelerate Video Diffusion Transformers (3 citations)
- Generating Attribute-Aware Human Motions from Textual Prompt (2 citations)
- GenVidBench: A 6-Million Benchmark for AI-Generated Video Detection (2 citations)
- HiTVideo: Hierarchical Tokenizers for Enhancing Text-to-Video Generation with Autoregressive Large Language Models (2 citations)
- OmniVDiff: Omni Controllable Video Diffusion for Generation and Understanding (2 citations)
- OwlCap: Harmonizing Motion-Detail for Video Captioning via HMD-270K and Caption Set Equivalence Reward (2 citations)
and 67 more in the interactive map.
Video-language understanding · 67 papers
- ReTrack: Evidence-Driven Dual-Stream Directional Anchor Calibration Network for Composed Video Retrieval (7 citations)
- StreamKV: Streaming Video Question-Answering with Segment-based KV Cache Retrieval and Compression (7 citations)
- VQAThinker: Exploring Generalizable and Explainable Video Quality Assessment via Reinforcement Learning (5 citations)
- Exo2Ego: Exocentric Knowledge Guided MLLM for Egocentric Video Understanding (3 citations)
- KTV: Keyframes and Key Tokens Selection for Efficient Training-Free Video LLMs (3 citations)
- Less Is More: Vision Representation Compression for Efficient Video Generation with Large Language Models (3 citations)
- ReasonAct: Progressive Training for Fine-Grained Video Reasoning in Small Models (3 citations)
- Scaling-up Perceptual Video Quality Assessment (3 citations)
- SpaceVLLM: Endowing Multimodal Large Language Model with Spatio-Temporal Video Grounding Capability (3 citations)
- VideoChat-A1: Thinking with Long Videos by Chain-of-Shot Reasoning (3 citations)
- VTD-CLIP: Video-to-Text Discretization via Prompting CLIP (3 citations)
- CogStream: Context-guided Streaming Video Question Answering (2 citations)
and 55 more in the interactive map.
Video segmentation and tracking · 63 papers
- Spatio-Temporal Context Learning with Temporal Difference Convolution for Moving Infrared Small Target Detection (8 citations)
- Ultralight Polarity-Split Neuromorphic SNN for Event-Stream Super-Resolution (4 citations)
- Exploiting Blurry Representations for Event-guided Video Super-Resolution (3 citations)
- HATIR: Heat-Aware Diffusion for Turbulent Infrared Video Super-Resolution (3 citations)
- Open-World Object Counting in Videos (3 citations)
- S2C: A Noise-Resistant Difference Learning Framework for Unsupervised Change Detection in VHR Remote Sensing Images (3 citations)
- DLVINet: Advancing Dual-Lens Video Inpainting Beyond Parallax Constraints (2 citations)
- Few-Shot Precise Event Spotting via Unified Multi-Entity Graph and Distillation (2 citations)
- Learning to Tell Apart: Weakly Supervised Video Anomaly Detection via Disentangled Semantic Alignment (2 citations)
- RAPTOR: Real-Time High-Resolution UAV Video Prediction with Efficient Video Attention (2 citations)
- RefineVAD: Semantic-Guided Feature Recalibration for Weakly Supervised Video Anomaly Detection (2 citations)
- Seeing the Unseen: Zooming in the Dark with Event Cameras (2 citations)
and 51 more in the interactive map.
Action recognition · 19 papers
- Gait Transformer: End-to-End Transformer Backbone for Gait Recognition (3 citations)
- FineTec: Fine-Grained Action Recognition Under Temporal Corruption via Skeleton Decomposition and Sequence Completion (2 citations)
- Human2Robot: Learning Robot Actions from Paired Human-Robot Videos (2 citations)
- What-Meets-Where: Unified Learning of Action and Contact Localization in Images (2 citations)
- Decompose and Conquer: Compositional Reasoning for Zero-Shot Temporal Action Localization (1 citation)
- Object-Centric Latent Action Learning (1 citation)
- Scene-Aware Spatiotemporal Generalization: Towards Robust Temporal Action Detection Across Domains (1 citation)
- SUGAR: Learning Skeleton Representation with Visual-Motion Knowledge for Action Recognition (1 citation)
- Whole-Field Action Sensing via Wearable Single-Channel EMG Sensors and Resource-Efficient Motion Network (1 citation)
- Backtrace Mamba: Reviving Critical Temporal Contexts via Hierarchical Memory Compression for Online Action Detection
- DeFB: Decomposed Feature Learning for Real-Time Multi-Person Eyeblink Detection in Untrimmed In-the-Wild Videos
- Explainable Depression Assessment from Face Videos by Weakly Supervised Learning
and 7 more in the interactive map.