List your company
ResearchFormerly WizWand

Action Recognition on HMDB51 (Mean Accuracy)

81.3Mean AccuracyOurs ViT-L
38.03649.26860.571.732Dec 18, 2020Oct 14, 2021Aug 11, 2022Jun 8, 2023Apr 4, 2024Jan 30, 2025Nov 27, 2025
Updated 8mo ago

Evaluation results

MethodLinks
Ours ViT-LModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning, Backbone=ViT-L2022.0781.3
Ours ViT-L (336↑)Modality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning, Backbone=ViT-L, Resolution=3362022.0781.3
TDNModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning2022.0776.4
TDNPretrain=ImageNet + Kinetics, Backbone=ResNet502020.1276.3
S3DPretrain=ImageNet+Kinetics, Backbone=Inception V22020.1275.9
S3D-GModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning2022.0775.9
MVFNetModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning2022.0775.7
I3DPretrain=ImageNet+Kinetics, Backbone=Inception V22020.1274.8
I3DModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning2022.0774.8
R(2+1)DPretrain=Kinetics, Backbone=ResNet342020.1274.5
R(2+1)DModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning2022.0774.5
TSMModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning2022.0773.5
TEAPretrain=ImageNet + Kinetics, Backbone=ResNet502020.1273.3
TSMPretrain=Kinetics, Backbone=ResNet502020.1273.2
STMPretrain=ImageNet + Kinetics, Backbone=ResNet502020.1272.2
STMModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning2022.0772.2
TEINetModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning2022.0772.1
ARTNetPretrain=Kinetics, Backbone=ResNet182020.1270.9
ARTNetModality=RGB, Pre-training=Kinetics, Transfer Protocol=Transfer Learning2022.0770.9
GA2-CLIPK=16, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1170.8
ViLT-CLIPK=16, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1169.6
GA2-CLIPK=8, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1168.1
ViFi-CLIPK=16, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1167.4
ViLT-CLIPK=8, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1166.9
ViFi-CLIPK=16, Adaptation Protocol=Tuning pre-trained image VL models2025.1166.8
CLIP text-FTK=16, Adaptation Protocol=Tuning pre-trained image VL models2025.1165
ViFi-CLIPK=8, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1164.8
ViFi-CLIPK=8, Adaptation Protocol=Tuning pre-trained image VL models2025.1164.5
GA2-CLIPK=4, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1164.3
X-CLIPK=16, Adaptation Protocol=Adapting pre-trained image VL models2025.1164
CLIP text-FTK=8, Adaptation Protocol=Tuning pre-trained image VL models2025.1163.1
X-CLIPK=8, Adaptation Protocol=Adapting pre-trained image VL models2025.1162.8
ViFi-CLIPK=4, Adaptation Protocol=Tuning pre-trained image VL models2025.1162.7
A5K=16, Adaptation Protocol=Adapting pre-trained image VL models2025.1162.4
CLIP image-FTK=16, Adaptation Protocol=Tuning pre-trained image VL models2025.1162
ViLT-CLIPK=4, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1161.9
GA2-CLIPK=2, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1161.9
CLIP text-FTK=4, Adaptation Protocol=Tuning pre-trained image VL models2025.1161.6
ViLT-CLIPK=2, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1160.6
ViFi-CLIPK=4, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1160
ActionCLIPK=16, Adaptation Protocol=Adapting pre-trained image VL models2025.1159.1
ActionCLIPK=4, Adaptation Protocol=Adapting pre-trained image VL models2025.1157.9
CLIP image-FTK=8, Adaptation Protocol=Tuning pre-trained image VL models2025.1157.8
ViFi-CLIPK=2, Adaptation Protocol=Prompt tuning pre-trained image VL models2025.1157.8
ActionCLIPK=8, Adaptation Protocol=Adapting pre-trained image VL models2025.1157.3
X-CLIPK=4, Adaptation Protocol=Adapting pre-trained image VL models2025.1157.3
ViFi-CLIPK=2, Adaptation Protocol=Tuning pre-trained image VL models2025.1157.2
A5K=8, Adaptation Protocol=Adapting pre-trained image VL models2025.1156
C3DPretrain=Sports-1M, Backbone=ResNet182020.1254.9
CLIP image-FTK=4, Adaptation Protocol=Tuning pre-trained image VL models2025.1154.9
CLIP text-FTK=2, Adaptation Protocol=Tuning pre-trained image VL models2025.1154.5
TSNPretrain=ImageNet, Backbone=Inception V22020.1253.7
X-CLIPK=2, Adaptation Protocol=Adapting pre-trained image VL models2025.1153
A5K=4, Adaptation Protocol=Adapting pre-trained image VL models2025.1150.7
CLIP image-FTK=2, Adaptation Protocol=Tuning pre-trained image VL models2025.1149.6
ActionCLIPK=2, Adaptation Protocol=Adapting pre-trained image VL models2025.1147.5
Vanilla CLIPK=2, Adaptation Protocol=Adapting pre-trained image VL models2025.1141.9
Vanilla CLIPK=4, Adaptation Protocol=Adapting pre-trained image VL models2025.1141.9
Vanilla CLIPK=8, Adaptation Protocol=Adapting pre-trained image VL models2025.1141.9
Vanilla CLIPK=16, Adaptation Protocol=Adapting pre-trained image VL models2025.1141.9
A5K=2, Adaptation Protocol=Adapting pre-trained image VL models2025.1139.7