List your company
ResearchFormerly WizWand

Phase Classification on Penn Action (test)

86.05AccuracySupervised Learning
42.505253.810165.11576.4199Apr 16, 2019Aug 13, 2019Dec 10, 2019Apr 7, 2020Aug 4, 2020Dec 1, 2020Mar 31, 2021
Updated 8mo ago

Evaluation results

MethodLinks
Supervised LearningBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=1.02019.0486.05———
TCC + TCNBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=1.02019.0485.22———
TCC + TCNBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.52019.0484.79———
LAV2021.0384.47———
TCCBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=1.02019.0484.45———
TCC + SaLBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=1.02019.0484.29———
TCC + TCNBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.12019.0484.27———
LAV + TCC2021.0384.14———
TCNBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=1.02019.0484.04———
TCC2021.0383.94———
TCNBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.52019.0483.67———
TCC + SaLBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.52019.0483.46———
TCCBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.52019.0483.35———
TCN2021.0382.99———
Supervised LearningBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.52019.0482.78———
SAL2021.0382.03———
TCNBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.12019.0481.99———
TCC + SaLBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.12019.0481.93———
TCCBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.12019.0481.26———
Supervised Learning% of Labels=1.0, Backbone=VGG-M, Training Protocol=from scratch2019.0479.98———
SaLBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=1.02019.0479.96———
LAVJoint all-action model=true, Shared single encoder=true, Classifier=13 SVMs2021.0378.68———
SaLBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.52019.0478.26———
TCC% of Labels=1.0, Backbone=VGG-M, Training Protocol=from scratch2019.0477.3———
TCC% of Labels=0.5, Backbone=VGG-M, Training Protocol=from scratch2019.0476.39———
SaLBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.12019.0474.87———
TCC% of Labels=0.1, Backbone=VGG-M, Training Protocol=from scratch2019.0474.68———
TCCJoint all-action model=true, Shared single encoder=true, Classifier=13 SVMs2021.0374.39———
Supervised Learning% of Labels=0.5, Backbone=VGG-M, Training Protocol=from scratch2019.0472.86———
SaL% of Labels=1.0, Backbone=VGG-M, Training Protocol=from scratch2019.0472.53———
TCN% of Labels=1.0, Backbone=VGG-M, Training Protocol=from scratch2019.0472.15———
TCN% of Labels=0.5, Backbone=VGG-M, Training Protocol=from scratch2019.0471.41———
SaL% of Labels=0.5, Backbone=VGG-M, Training Protocol=from scratch2019.0471.1———
TCN% of Labels=0.1, Backbone=VGG-M, Training Protocol=from scratch2019.0469.65———
SALJoint all-action model=true, Shared single encoder=true, Classifier=13 SVMs2021.0368.15———
TCNJoint all-action model=true, Shared single encoder=true, Classifier=13 SVMs2021.0368.09———
Supervised LearningBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.12019.0467.1———
SaL% of Labels=0.1, Backbone=VGG-M, Training Protocol=from scratch2019.0466.15———
ImageNet FeaturesBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=1.02019.0452.86———
ImageNet FeaturesBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.52019.0450.91———
Supervised Learning% of Labels=0.1, Backbone=VGG-M, Training Protocol=from scratch2019.0450.71———
Random FeaturesBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=1.02019.0446.81———
Random FeaturesBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.52019.0446.19———
ImageNet FeaturesBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.12019.0444.96———
Random FeaturesBackbone=ResNet-50, Pre-trained=ImageNet, Evaluation Protocol=Fine-tuning, % of Labels=0.12019.0444.18———
CASAPose=true, ImageNet pre-trained=false2022.04—88.5591.8792.2
LAVPose=false, ImageNet pre-trained=true2022.04—83.5683.9584.25
LAVPose=true, ImageNet pre-trained=false2022.04—79.8380.280.2
SaLPose=false, ImageNet pre-trained=true2022.04—74.8778.2679.96
TCCPose=false, ImageNet pre-trained=true2022.04—79.7281.1181.35
TCCPose=true, ImageNet pre-trained=false2022.04—79.5383.7584.51
TCNPose=false, ImageNet pre-trained=true2022.04—81.9983.6784.04