Self Supervised Action Recognition On Hmdb51

评估指标

Frozen

Pre-Training Dataset

Top-1 Accuracy

评测结果

各个模型在此基准测试上的表现结果

				Paper Title	Repository
MVD (ViT-B)	false	Kinetics400	79.7	Masked Video Distillation: Rethinking Masked Feature Modeling for Self-supervised Video Representation Learning
M3Video	false	Kinetics400	78.0	Masked Motion Encoding for Self-Supervised Video Representation Learning
pBYOL	false	Kinetics400	75.0	A Large-Scale Study on Unsupervised Spatiotemporal Representation Learning
SCE (R3D-50)	false	Kinetics400	74.7	Similarity Contrastive Estimation for Image and Video Soft Contrastive Self-Supervised Learning
VideoMAE	false	Kinetics400	73.3	VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training
BraVe:V-FA (TSM-50x2)	false	-	70.5	Broaden Your Views for Self-Supervised Video Learning
CVRL (R3D-152 2x; K600)	false	Kinetics600	69.9	Spatiotemporal Contrastive Video Representation Learning
XKD (ViT-B/112/16)	-	-	69	XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning
XDC	false	IG-Kinetics	68.9	Self-Supervised Learning by Cross-Modal Audio-Video Clustering
CVRL (R3D-50; K600)	false	Kinetics600	68.0	Spatiotemporal Contrastive Video Representation Learning
CrissCross (AudioSet)	false	AudioSet	66.8	Self-Supervised Audio-Visual Representation Learning with Relaxed Cross-Modal Synchronicity
CVRL (R3D-50; K400)	false	Kinetics400	66.7	Spatiotemporal Contrastive Video Representation Learning
XDC	false	IG-Random	66.5	Self-Supervised Learning by Cross-Modal Audio-Video Clustering
XKD-Modality-Agnostic (ViT-B/112/16)	-	-	65.9	XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning
VideoMS (ViT-B)	false	no extra data	65.8	EVEREST: Efficient Masked Video Autoencoder by Removing Redundant Spatiotemporal Tokens
RSPNet	false	Kinetics400	64.7	RSPNet: Relative Speed Perception for Unsupervised Video Representation Learning
CrissCross (Kinetics400)	false	Kinetics400	64.7	Self-Supervised Audio-Visual Representation Learning with Relaxed Cross-Modal Synchronicity
AVID+CMA (Modified R2+1D-18 on Audioset)	false	Audioset (Video+Audio)	64.7	Audio-Visual Instance Discrimination with Cross-Modal Agreement
ELo	false	-	64.5	Evolving Losses for Unsupervised Video Representation Learning	-
AVID (Modified R2+1D-18 on Audioset)	false	Audioset (Video+Audio)	64.1	Audio-Visual Instance Discrimination with Cross-Modal Agreement

0 of 48 row(s) selected.