HyperAI초신경

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

한국어

HyperAI초신경

Self Supervised Action Recognition On Ucf101 1

평가 지표

3-fold Accuracy

Pretrain

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	3-fold Accuracy	Pretrain	Paper Title	Repository
CVRL (R3D-50; K400)	92.2	K400	Spatiotemporal Contrastive Video Representation Learning	-
ViCC (R2+1D; RGB)	82.8	UCF101	Self-supervised Video Representation Learning with Cross-Stream Prototypical Contrasting	-
AVID	91.5	-	Audio-Visual Instance Discrimination with Cross-Modal Agreement	-
CoCLR	87.9	K400	Self-supervised Co-training for Video Representation Learning	-
ViCC (R2+1D; R+F)	88.8	UCF101	Self-supervised Video Representation Learning with Cross-Stream Prototypical Contrasting	-
CVRL (R3D-50; K600)	93.4	K600	Spatiotemporal Contrastive Video Representation Learning	-
ViCC (S3D; R+F)	90.5	UCF101	Self-supervised Video Representation Learning with Cross-Stream Prototypical Contrasting	-
CVRL (R3D-152 2x; K600)	93.9	K600	Spatiotemporal Contrastive Video Representation Learning	-
AVTS	89.0	-	Cooperative Learning of Audio and Video Models from Self-Supervised Synchronization	-
BraVe:V-FA (TSM-50x2)	95.7	-	Broaden Your Views for Self-Supervised Video Learning	-
ViCC (S3D; RGB)	84.3	UCF101	Self-supervised Video Representation Learning with Cross-Stream Prototypical Contrasting	-
MMV	91.5	-	Self-Supervised MultiModal Versatile Networks	-
ELo	93.8	-	Evolving Losses for Unsupervised Video Representation Learning	-
XDC	95.5	-	Self-Supervised Learning by Cross-Modal Audio-Video Clustering	-

0 of 14 row(s) selected.

Self Supervised Action Recognition On Ucf101 1 | SOTA | HyperAI초신경