Natural Language Moment Retrieval on MAD

8개 결과 · ⬇ CSV · JSON

R@1,IoU=0.1

Rank	Model	R@1,IoU=0.1	R@1,IoU=0.3	R@1,IoU=0.5	R@10,IoU=0.1	R@10,IoU=0.3	R@10,IoU=0.5	R@100,IoU=0.1	R@100,IoU=0.3	Paper	Code	Year
1	ReVisionLLM	17.3	12.7	6.7	–	–	–	–	–	ReVisionLLM: Recursive Vision-Language Model for Temporal Grounding in Hour-Long Videos	tanveer81/revisionllm	2024
2	DeCafNet	13.25	10.96	7.06	–	–	–	–	–	DeCafNet: Delegate and Conquer for Efficient Temporal Grounding in Long Videos	zijialewislu/cvpr2025-decafnet	2025
3	RGNet	12.43	9.48	5.61	–	–	–	–	–	RGNet: A Unified Clip Retrieval and Grounding Network for Long Videos	tanveer81/revisionllm · tanveer81/rgnet	2023
4	Zero-Shot CLIP + Guidance Model	9.3	4.65	2.16	24.30	17.73	11.09	47.35	39.58	Localizing Moments in Long Video Via Multimodal Guidance	waybarrios/guidance-based-video-grounding	2023
5	CLIP	6.57	3.13	1.39	20.26	14.13	8.38	47.73	36.98	MAD: A Scalable Dataset for Language Grounding in Videos from Movie Audio Descriptions	Soldelli/MAD	2021
6	VLG-Net + Guidance Model	5.60	4.28	2.48	23.64	19.86	13.72	55.59	49.38	Localizing Moments in Long Video Via Multimodal Guidance	waybarrios/guidance-based-video-grounding	2023
7	VLG-Net	3.50	2.63	1.61	18.32	15.2	10.18	49.65	43.95	MAD: A Scalable Dataset for Language Grounding in Videos from Movie Audio Descriptions	Soldelli/MAD	2021
8	Random Chance	0.09	0.04	0.01	0.88	0.39	0.14	8.47	3.80	MAD: A Scalable Dataset for Language Grounding in Videos from Movie Audio Descriptions	Soldelli/MAD	2021