HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Video Retrieval On Fivr 200K

평가 지표

mAP (CSVR)

mAP (DSVR)

mAP (ISVR)

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	mAP (CSVR)	mAP (DSVR)	mAP (ISVR)	Paper Title	Repository
ViSiLf	0.797	0.843	0.660	ViSiL: Fine-grained Spatio-Temporal Video Similarity Learning
TCAc	0.553	0.570	0.473	Temporal Context Aggregation for Video Retrieval with Contrastive Learning
DnS (S^f_B)	0.863	0.909	0.729	DnS: Distill-and-Select for Efficient and Accurate Video Indexing and Retrieval
ViSiLsym	0.792	0.833	0.654	ViSiL: Fine-grained Spatio-Temporal Video Similarity Learning
VRAG (CS)	0.678	0.723	0.554	VRAG: Region Attention Graphs for Content-Based Video Retrieval	-
ViSiLv (pt)	0.854	0.899	0.723	ViSiL: Fine-grained Spatio-Temporal Video Similarity Learning
DnS (S^c)	0.558	0.574	0.476	DnS: Distill-and-Select for Efficient and Accurate Video Indexing and Retrieval
S2VS	0.879	0.927	0.746	Self-Supervised Video Similarity Learning
Jo et al. (SCFV+TNIP)	0.833	0.896	0.674	Exploring the Temporal Cues to Enhance Video Retrieval on Standardized CDVA
TCAf	0.830	0.877	0.703	Temporal Context Aggregation for Video Retrieval with Contrastive Learning
TCAsym	0.698	0.728	0.592	Temporal Context Aggregation for Video Retrieval with Contrastive Learning
ViSiLv (tf)	0.841	0.892	0.702	ViSiL: Fine-grained Spatio-Temporal Video Similarity Learning
VVS	0.689	0.711	0.590	VVS: Video-to-Video Retrieval with Irrelevant Frame Suppression
S2VS	0.878	0.925	0.739	Self-Supervised Video Similarity Learning
DnS (S^f_A)	0..875	0.921	0.741	DnS: Distill-and-Select for Efficient and Accurate Video Indexing and Retrieval
VRAG (video)	0.470	0.484	0.399	VRAG: Region Attention Graphs for Content-Based Video Retrieval	-
LAMV	0.466	0.496	0.371	LAMV: Learning to Align and Match Videos With Kernelized Temporal Layers

0 of 17 row(s) selected.