List your company
ResearchFormerly WizWand

Video Action Recognition on HMDB-51 (Mean class accuracy)

75.9Mean Class AccuracyS3D-G
38.25248.02657.867.574Dec 13, 2020Apr 12, 2021Aug 11, 2021Dec 9, 2021Apr 9, 2022Aug 7, 2022Dec 6, 2022
Updated 8mo ago

Evaluation results

MethodLinks
S3D-GBackbone=Inception V1, Pre-train=Kinetics, Modality=RGB, Splits=32020.1275.9
MVFNetBackbone=ResNet-50, Pre-train=Kinetics, Modality=RGB, Frames=16, Splits=32020.1275.7
I3DBackbone=Inception V1, Pre-train=Kinetics, Modality=RGB, Splits=32020.1274.8
R(2+1)DBackbone=Inception V1, Pre-train=Kinetics, Modality=RGB, Splits=32020.1274.5
TSMBackbone=ResNet-50, Pre-train=Kinetics, Modality=RGB, Splits=32020.1273.5
ECO EnBackbone=BNIncep+Res3D-18, Pre-train=Kinetics, Modality=RGB, Splits=32020.1272.4
STMBackbone=ResNet-50, Pre-train=Kinetics, Modality=RGB, Splits=32020.1272.2
TEINetBackbone=ResNet-50, Pre-train=Kinetics, Modality=RGB, Splits=32020.1272.1
ARTNetBackbone=ResNet-18, Pre-train=Kinetics, Modality=RGB, Splits=32020.1270.9
ViFi-CLIPK=16, Adaptation strategy=Tuning pre-trained image VL models2022.1266.8
CLIP text-FTK=16, Adaptation strategy=Tuning pre-trained image VL models2022.1265
ViFi-CLIPK=8, Adaptation strategy=Tuning pre-trained image VL models2022.1264.5
XCLIPK=16, Adaptation strategy=Adapting pre-trained image VL models2022.1264
CLIP text-FTK=8, Adaptation strategy=Tuning pre-trained image VL models2022.1263.1
XCLIPK=8, Adaptation strategy=Adapting pre-trained image VL models2022.1262.8
ViFi-CLIPK=4, Adaptation strategy=Tuning pre-trained image VL models2022.1262.7
A5K=16, Adaptation strategy=Adapting pre-trained image VL models2022.1262.4
CLIP image-FTK=16, Adaptation strategy=Tuning pre-trained image VL models2022.1262
CLIP text-FTK=4, Adaptation strategy=Tuning pre-trained image VL models2022.1261.6
ActionCLIPK=16, Adaptation strategy=Adapting pre-trained image VL models2022.1259.1
ActionCLIPK=4, Adaptation strategy=Adapting pre-trained image VL models2022.1257.9
CLIP image-FTK=8, Adaptation strategy=Tuning pre-trained image VL models2022.1257.8
ActionCLIPK=8, Adaptation strategy=Adapting pre-trained image VL models2022.1257.3
XCLIPK=4, Adaptation strategy=Adapting pre-trained image VL models2022.1257.3
ViFi-CLIPK=2, Adaptation strategy=Tuning pre-trained image VL models2022.1257.2
A5K=8, Adaptation strategy=Adapting pre-trained image VL models2022.1256
CLIP image-FTK=4, Adaptation strategy=Tuning pre-trained image VL models2022.1254.9
CLIP text-FTK=2, Adaptation strategy=Tuning pre-trained image VL models2022.1254.5
XCLIPK=2, Adaptation strategy=Adapting pre-trained image VL models2022.1253
A5K=4, Adaptation strategy=Adapting pre-trained image VL models2022.1250.7
CLIP image-FTK=2, Adaptation strategy=Tuning pre-trained image VL models2022.1249.6
ActionCLIPK=2, Adaptation strategy=Adapting pre-trained image VL models2022.1247.5
Vanilla CLIPK=2, Adaptation strategy=Adapting pre-trained image VL models2022.1241.9
Vanilla CLIPK=4, Adaptation strategy=Adapting pre-trained image VL models2022.1241.9
Vanilla CLIPK=8, Adaptation strategy=Adapting pre-trained image VL models2022.1241.9
Vanilla CLIPK=16, Adaptation strategy=Adapting pre-trained image VL models2022.1241.9
A5K=2, Adaptation strategy=Adapting pre-trained image VL models2022.1239.7