SOTAVerified

Video Understanding

A crucial task of Video Understanding is to recognise and localise (in space and time) different actions or events appearing in the video.

Source: Action Detection from a Robot-Car Perspective

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 51–75 of 1149 papers

TitleStatusHype
Threading Keyframe with Narratives: MLLMs as Strong Long Video Comprehenders—0
SiLVR: A Simple Language-based Video Reasoning FrameworkCode1
Learning reusable concepts across different egocentric video understanding tasks—0
VUDG: A Dataset for Video Understanding Domain Generalization—0
VideoCAD: A Large-Scale Video Dataset for Learning UI Interactions and 3D Reasoning from CAD SoftwareCode1
Time Blindness: Why Video-Language Models Can't See What Humans Can?—0
ScaleLong: A Multi-Timescale Benchmark for Long Video UnderstandingCode0
Multi-RAG: A Multimodal Retrieval-Augmented Generation System for Adaptive Video Understanding—0
MaCP: Minimal yet Mighty Adaptation via Hierarchical Cosine Projection—0
VideoREPA: Learning Physics for Video Generation through Relational Alignment with Foundation ModelsCode2
PreFM: Online Audio-Visual Event Parsing via Predictive Future ModelingCode1
One Trajectory, One Token: Grounded Video Tokenization via Panoptic Sub-object TrajectoryCode2
VideoReasonBench: Can MLLMs Perform Vision-Centric Complex Video Reasoning?Code1
Universal Visuo-Tactile Video Understanding for Embodied Interaction—0
VidText: Towards Comprehensive Evaluation for Video Text UnderstandingCode1
MUSEG: Reinforcing Video Temporal Understanding via Timestamp-Aware Multi-Segment GroundingCode1
Two Causally Related Needles in a Video Haystack—0
AdaTP: Attention-Debiased Token Pruning for Video Large Language Models—0
TUNA: Comprehensive Fine-grained Temporal Understanding Evaluation on Dense Dynamic VideosCode0
Sparse-to-Dense: A Free Lunch for Lossless Acceleration of Video Understanding in LLMs—0
Deep Video Discovery: Agentic Search with Tool Use for Long-form Video Understanding—0
Four Eyes Are Better Than Two: Harnessing the Collaborative Potential of Large Models via Differentiated Thinking and Complementary Ensembles—0
SoccerChat: Integrating Multimodal Data for Enhanced Soccer Game UnderstandingCode0
QuickVideo: Real-Time Long Video Understanding with System Algorithm Co-DesignCode2
Fact-R1: Towards Explainable Video Misinformation Detection with Deep ReasoningCode1
Show:102550
← PrevPage 3 of 46Next →

No leaderboard results yet.