List your company
ResearchFormerly WizWand

Facial Expression Recognition on FERPlus (test)

0.9555AccuracyPAtt-Lite
0.826020.8596350.893250.926865Apr 29, 2018May 30, 2019Jun 30, 2020Aug 1, 2021Sep 2, 2022Oct 4, 2023Nov 4, 2024
Updated 8mo ago

Evaluation results

MethodLinks
PAtt-Lite# Params=1.10M2023.060.9555
CIAO# Params=17.9M2023.060.945
ARBEx# Params=-2023.060.9309
GReFELVenue=Current Paper2024.100.9309
POSTER++Venue='232024.100.9228
QCSPre-trained=AffectNet-82024.110.9185
LA-NetVenue=ICCV'232024.100.9178
POSTER# Params=71.8M2023.060.9162
POSTERVenue=IC-W'222024.100.9162
POSTERyear=20232024.110.9162
DCS2024.110.9141
QCS2024.110.9141
S2D w/o TMAPre-trained on=AffectNet-82023.120.9117
S2Dyear=2024, Pre-trained=AffectNet-82024.110.9117
PF-ViT2023.120.9116
PF-VITbackbone=ViT-B, pre-training data=AffectNet-270K, venue=TCSS'242022.070.9116
FG-AGRbackbone=Hybrid ViT, pre-training data=MS-Celeb-1M, venue=TCSVT'232022.070.9109
S2D w/o TMA2023.120.9101
TAN2023.120.91
TANyear=20232024.110.91
APViTTransformer-based=true2022.120.9086
APViT# Params=65.2M2023.060.9086
APViT2023.120.9086
APVITyear=20222024.110.9086
TransFERTransformer-based=true2022.120.9083
TransFER# Params=65.2M2023.060.9083
TransFER2023.120.9083
TransFERbackbone=Hybrid ViT, pre-training data=MS-Celeb-1M, venue=ICCV'212022.070.9083
TransFERyear=20212024.110.9083
DDAMFN# Params=4.11M2023.060.9074
KTNTransformer-based=false2022.120.9049
MTACbackbone=Swin-S [58], pre-training data=ImageNet1K, venue=TMM'242022.070.904
MAE-VITbackbone=ViT-B, pre-training data=AffectNet-270K, venue=TCSS'242022.070.9018
EAC# Params=11.2M2023.060.8964
EAC2023.120.8964
EACbackbone=ResNet-18, pre-training data=MS-Celeb-1M, venue=ECCV'222022.070.8964
EACyear=20222024.110.8964
MRANbackbone=Hybrid ViT, pre-training data=MS-Celeb-1M, venue=TCSVT'232022.070.8959
DMUEBackbone=ResNet-50IBN2021.040.8951
DMUEBackbone=ResNet-50IBN2021.040.8951
DMUE2023.120.8951
DMUEbackbone=Res-50IBN, pre-training data=Ms-Celeb-1M, venue=CVPR'212022.070.8951
DMUEyear=20212024.110.8951
SCAN-CCI# Params=70M2023.060.8942
SCNTransformer-based=false2022.120.8935
SCN2023.120.8935
SCNbackbone=ResNet-18, pre-training data=Ms-Celeb-1M, venue=CVPR'202022.070.8935
MAE-VITbackbone=ViT-S, pre-training data=AffectNet-270K, venue=TCSS'242022.070.8935
MViTTransformer-based=true2022.120.8922
MVT2023.120.8922
MVTbackbone=DeiT-S [44], pre-training data=ImageNet1K, venue=Arxiv'212022.070.8922
RAN-VGG16Network=VGG16, Pre-trained Dataset=VGG_Face, Label distribution supervision=true2019.050.8916
RAN-VGG16Transformer-based=false2022.120.8916
RAN# Params=11.2M2023.060.8916
VTFFTransformer-based=true2022.120.8881
VTFF# Params=80.1M2023.060.8881
VTFF2023.120.8881
VTFFbackbone=Hybrid ViT, pre-training data=MS-Celeb-1M, venue=TAFFC'232022.070.8881
[1]Network=SeNet50, Pre-trained Dataset=VGG_Face2, Label distribution supervision=true2019.050.888
SeNet50Data Annotation Type=extra distribution data2021.040.888
SeNet50Backbone=SeNet50, Extra distribution data=true2021.040.888
SeNet50Transformer-based=false2022.120.888
MAE-VITbackbone=ViT-T, pre-training data=AffectNet-270K, venue=TCSS'242022.070.8867
DMUEBackbone=ResNet-182021.040.8864
DMUEBackbone=ResNet-182021.040.8864
DMUEbackbone=ResNet-18, pre-training data=Ms-Celeb-1M, venue=CVPR'212022.070.8864
RAN-ResNet18Network=ResNet18, Pre-trained Dataset=MS_Celeb_1M2019.050.8855
RAN2021.040.8855
RAN2021.040.8855
RANTransformer-based=false2022.120.8855
IPD-FERbackbone=ResNet-18, pre-training data=CASIA-WebFace [53], venue=TAFFC'222022.070.8842
SCN2021.040.8801
SCN2021.040.8801
SCNyear=20202024.110.8801
CNNs and BOVW + local SVMData Augmentation (aug.)=false2018.040.8776
[27]Network=ResNet18+VGG16, Pre-trained Dataset=None2019.050.874
Res+VGG2021.040.874
Res+VGGBackbone=ResNet + VGG2021.040.874
CNNs and BOVW + local SVMData Augmentation (aug.)=true2018.040.8725
ESR-9Ensemble size=92020.010.8715
EACVenue=ECCV'222024.100.8703
CNNs and BOVW + global SVMData Augmentation (aug.)=true2018.040.8696
CNNs and BOVW + global SVMData Augmentation (aug.)=false2018.040.8668
PTTransformer-based=false2022.120.866
SHCNN2020.010.8654
SCNVenue=CVPR'202024.100.8597
[5]Network=VGG13, Pre-trained Dataset=None, Label distribution supervision=true2019.050.851
PLDData Annotation Type=extra distribution data2021.040.851
PLDExtra distribution data=true2021.040.851
PLDTransformer-based=false2022.120.851
VGG16-PLDLoss/Label=Probabilistic Label Distribution2020.010.8499
Barsoum et al.Data Augmentation (aug.)=true2018.040.8499
VGG16-CELLoss/Label=Cross-Entropy Loss2020.010.8472
TFE-JL2020.010.843
TFE-joint learningTransformer-based=false2022.120.8429
VGG16-MLAggregator=ML2020.010.8397
VGG16-MVAggregator=Majority Voting2020.010.8385
RANVenue=TIP'202024.100.8363
ResNet18 + FCStructure=ResNet18 with Fully Connected layer2020.010.834
ResNet18Backbone=ResNet-182020.010.831