List your company
ResearchFormerly WizWand

Referring Image Segmentation on G-Ref Google split (val)

65.79IoUCGFormer
33.269241.712150.15558.5979Nov 21, 2021Mar 9, 2022Jun 25, 2022Oct 12, 2022Jan 28, 2023May 16, 2023Sep 2, 2023
Updated 8mo ago

Evaluation results

MethodLinks
CGFormer2023.0965.79—62.51
VLT [6]Visual Encoder=Swin-B, Textual Encoder=BERT2023.0662.8——
VLTVisual Backbone=Swin-B, Textual Encoder=BERT2022.1062.873.84—
ReLAVisual Encoder=Swin-B, Textual Encoder=BERT2023.0662.7——
MaILResolution=Fixed aspect ratio2021.1161.81——
MaILVisual Backbone=ViLT, Textual Encoder=BERT, large-scale pre-training=true2022.1061.81——
LAVTLanguage Model=BERT2021.1260.5——
LAVTVisual Encoder=Swin-B, Textual Encoder=BERT2023.0660.5——
LAVTVisual Backbone=Swin-B, Textual Encoder=BERT2022.1060.5——
CRIS2023.0960.36——
MaILResolution=416x4162021.1160.26——
ReSTRVisual Encoder=ViT-B, Textual Encoder=Transformer2023.0654.48——
ReSTRVisual Backbone=ViT-B, Textual Encoder=Transformer2022.1054.48——
VLT [5]Visual Encoder=Darknet53, Textual Encoder=bi-GRU2023.0652.02——
VLTVisual Backbone=Darknet53, Textual Encoder=bi-GRU2022.1052.0257.88—
EFN2021.1151.93——
EFNLanguage Model=Bi-GRU2021.1251.93——
EFNVisual Backbone=ResNet101, Textual Encoder=bi-GRU2022.1051.93——
BUSNetLanguage Model=Self-Att2021.1250.56——
BUSNetVisual Backbone=DeepLab-R101, Textual Encoder=Self-Att2022.1050.56——
ISFPVisual Backbone=Darknet53, Textual Encoder=Bi-GRU2022.1050.08——
BUSNet2021.1149.98——
CMPC+Language Model=LSTM2021.1249.89——
CMPC+Visual Backbone=DeepLab-R101, Textual Encoder=LSTM2022.1049.89——
VLT2021.1149.76——
VLTLanguage Model=Bi-GRU2021.1249.76——
VLT2023.0949.76——
CMPCLanguage Model=LSTM2021.1249.05——
JRNet2021.1148.95——
LSCM2021.1148.05——
LSCMLanguage Model=LSTM2021.1248.05——
LSCMVisual Backbone=DeepLab-R101, Textual Encoder=LSTM2022.1048.05——
BRINet2021.1148.04——
BRINetLanguage Model=LSTM2021.1248.04——
BRINetVisual Backbone=DeepLab-R101, Textual Encoder=LSTM2022.1048.04——
CGAN2021.1146.54——
CGANLanguage Model=Bi-GRU2021.1246.54——
CGAN2023.0946.54——
CGANVisual Backbone=DeepLab-R101, Textual Encoder=bi-GRU2022.1046.54——
STEP2021.1146.4——
STEPLanguage Model=Bi-LSTM2021.1246.4——
STEPVisual Backbone=DeepLab-R101, Textual Encoder=bi-LSTM2022.1046.4——
CAC2021.1144.32——
CACLanguage Model=Bi-LSTM2021.1244.32——
CACVisual Backbone=ResNet101, Textual Encoder=bi-LSTM2022.1044.32——
ASGN2021.1141.36——
CMSA2021.1139.98——
CMPC2021.1139.98——
CMSALanguage Model=None2021.1239.98——
CMSAVisual Backbone=DeepLab-R101, Textual Encoder=None2022.1039.98——
CMPCVisual Backbone=DeepLab-R101, Textual Encoder=LSTM2022.1039.98——
DMN2021.1136.76——
DMNLanguage Model=SRU2021.1236.76——
DMNVisual Backbone=DPN92, Textual Encoder=SRU2022.1036.76——
RRN2021.1136.45——
RRNLanguage Model=LSTM2021.1236.45——
RRNVisual Backbone=DeepLab-R101, Textual Encoder=LSTM2022.1036.45——
RMI2021.1134.52——
BUSNet2023.09——50.56
CMPC2023.09——49.05
CMSA2023.09——39.98
LAVT2023.09——60.5
LSCM2023.09——48.05
RRN2023.09——36.45