Moment Retrieval On Charades Sta

评估指标

R@1 IoU=0.5

R@1 IoU=0.7

评测结果

各个模型在此基准测试上的表现结果

			Paper Title
SG-DETR (w/ PT)	71.10	52.80	Saliency-Guided DETR for Moment Retrieval and Highlight Detection
LLaVA-MR	70.65	49.58	LLaVA-MR: Large Language-and-Vision Assistant for Video Moment Retrieval
FlashVTG	70.32	49.87	FlashVTG: Feature Layering and Adaptive Score Handling Network for Video Temporal Grounding
SG-DETR	70.20	49.50	Saliency-Guided DETR for Moment Retrieval and Highlight Detection
InternVideo2-6B	70.03	48.95	InternVideo2: Scaling Foundation Models for Multimodal Video Understanding
InternVideo2-1B	68.36	45.03	InternVideo2: Scaling Foundation Models for Multimodal Video Understanding
VideoChat-T (FT)	67.1	43.0	TimeSuite: Improving MLLMs for Long Video Understanding via Grounded Tuning
UniMD+Sync.	63.98	44.46	UniMD: Towards Unifying Moment Retrieval and Temporal Action Detection
LD-DETR	62.58	41.56	LD-DETR: Loop Decoder DEtection TRansformer for Video Moment Retrieval and Highlight Detection
VideoLights-B-pt	61.96	41.05	VideoLights: Feature Refinement and Cross-Task Alignment Transformer for Joint Video Highlight Detection and Moment Retrieval
UnLoc-L	60.8	38.4	UnLoc: A Unified Framework for Video Localization Tasks
BAM-DETR	59.95	39.38	BAM-DETR: Boundary-Aligned Moment Detection Transformer for Temporal Sentence Grounding in Videos
BM-DETR	59.48	38.33	Background-aware Moment Detection for Video Moment Retrieval
UVCOM	59.25	36.64	Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection
CG-DETR	58.44	36.34	Correlation-Guided Query-Dependency Calibration for Video Temporal Grounding
LLMEPET	58.31	36.49	Prior Knowledge Integration via LLM Encoding and Pseudo Event Regulation for Video Moment Retrieval
UnLoc-B	58.1	35.4	UnLoc: A Unified Framework for Video Localization Tasks
QD-DETR (Only Video)	57.31	32.55	Query-Dependent Video Representation for Moment Retrieval and Highlight Detection
video-mamba-suite	57.18	36.05	Video Mamba Suite: State Space Model as a Versatile Alternative for Video Understanding
Moment-DETR w/ PT (on 10K HowTo100M videos)	55.65	34.17	QVHighlights: Detecting Moments and Highlights in Videos via Natural Language Queries

0 of 25 row(s) selected.