List your company
ResearchFormerly WizWand

Action Recognition on HMDB (Top-1 Accuracy)

75.9Top-1 AccuracyS3D (supervised learning)
21.71635.78349.8563.917Mar 31, 2022
Updated 8mo ago

Evaluation results

MethodLinks
S3D (supervised learning)Model (Backbone)=S3D, Frozen=✖2022.0375.9
VT-TWINSPre-training Dataset=HTM, MM (Modality)=Text, Model (Backbone)=S3D, Frozen=✔2022.0357.9
MIL-NCEPre-training Dataset=HTM, MM (Modality)=Text, Model (Backbone)=I3D, Frozen=✔2022.0354.8
MIL-NCEPre-training Dataset=HTM, MM (Modality)=Text, Model (Backbone)=S3D, Frozen=✔2022.0353.1
AVTSPre-training Dataset=K600, MM (Modality)=Audio, Model (Backbone)=I3D, Frozen=✖2022.0353
CBTPre-training Dataset=K600, MM (Modality)=✖, Model (Backbone)=S3D, Frozen=✖2022.0344.6
3DRotNetPre-training Dataset=K600, MM (Modality)=✖, Model (Backbone)=S3D, Frozen=✖2022.0340
Shuffle & LearnPre-training Dataset=K600, MM (Modality)=✖, Model (Backbone)=S3D, Frozen=✖2022.0335.8
DPCPre-training Dataset=K400, MM (Modality)=✖, Model (Backbone)=3D-R34, Frozen=✖2022.0335.7
3D ST-puzzlePre-training Dataset=K400, MM (Modality)=✖, Model (Backbone)=3D-R18, Frozen=✖2022.0333.7
Wang et al.Pre-training Dataset=K400, MM (Modality)=Flow, Model (Backbone)=C3D, Frozen=✖2022.0333.4
Fernanado et al.Pre-training Dataset=UCF, MM (Modality)=✖, Model (Backbone)=AlexNet, Frozen=✖2022.0332.5
ClipOrderPre-training Dataset=UCF, MM (Modality)=✖, Model (Backbone)=R(2+1)D, Frozen=✖2022.0330.9
CBTPre-training Dataset=K600, MM (Modality)=✖, Model (Backbone)=S3D, Frozen=✔2022.0329.5
CMCPre-training Dataset=UCF, MM (Modality)=Flow, Model (Backbone)=CaffeNet, Frozen=✖2022.0326.7
GeometryPre-training Dataset=UCF, MM (Modality)=Flow, Model (Backbone)=CaffeNet, Frozen=✖2022.0326.7
OPNPre-training Dataset=UCF, MM (Modality)=✖, Model (Backbone)=VGG, Frozen=✖2022.0323.8