List your company
ResearchFormerly WizWand

Open Vocabulary Semantic Segmentation on COCO Object with background

47.4mIoUVIP
3.61614.98326.3537.717Nov 14, 2024Feb 12, 2025May 14, 2025Aug 13, 2025Nov 11, 2025Feb 10, 2026May 12, 2026
Updated 5mo ago

Evaluation results

MethodLinks
VIPExtra Backbone=X, Vision Model Configuration=Single vision model2026.0547.4
CorrCLIPExtra Backbone=DINO+SAM2, Vision Model Configuration=Multi vision models2026.0543.7
CLIPtrasePub. & Year=ECCV’24, Extra data=✗, Extra backbone=✗, Training free=✓2026.0343.6
TridentTrain=false, Backbone=ViT-H/142024.1142.2
TridentExtra Backbone=DINO+SAM, Vision Model Configuration=Multi vision models2026.0541.1
TridentTrain=false, Backbone=ViT-B/162024.1141.1
CLIPerExtra Backbone=Stable Diffusion, Vision Model Configuration=Multi vision models2026.0540
ProxyCLIPTrain=false, Backbone=ViT-H/142024.1138.6
FSAExtra Backbone=DINO, Vision Model Configuration=Multi vision models2026.0538
SFPExtra Backbone=X, Vision Model Configuration=Single vision model2026.0537.9
CASSExtra Backbone=DINO, Vision Model Configuration=Multi vision models2026.0537.8
CASSTrain=False, External Model=DINO, DS Hyperparameter=True, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0337.8
SC-CLIPExtra Backbone=X, Vision Model Configuration=Single vision model2026.0537.7
ProxyCLIP + GLATrain=False, External Model=DINO, DS Hyperparameter=False, Image Resize=448, Crop size=336, Stride=112, Rename trick=With rename trick on Background Classes2026.0337.7
ProxyCLIP + GLA (Best)Train=False, External Model=DINO, DS Hyperparameter=True, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0337.7
ProxyCLIP + GLATrain=False, External Model=DINO, DS Hyperparameter=False, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0337.7
ProxyCLIPExtra Backbone=DINO, Vision Model Configuration=Multi vision models2026.0537.5
ProxyCLIPTrain=false, Backbone=ViT-B/162024.1137.5
ProxyCLIPTrain=False, External Model=DINO, DS Hyperparameter=False, Image Resize=448, Crop size=336, Stride=112, Rename trick=With rename trick on Background Classes2026.0337.5
TTDTrain=true2024.1137.4
ProxyCLIP*Pub. & Year=ECCV’24, Extra data=✗, Extra backbone=DINOv2† (ViT-B/14), Training free=✓2026.0337.4
DIH-CLIPTrain=False, External Model=None, DS Hyperparameter=False, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0337.4
PEARLPub. & Year=current, Extra data=✗, Extra backbone=✗, Training free=✓2026.0337.3
ProxyCLIP + GLATrain=False, External Model=DINO, DS Hyperparameter=False, Image Resize=448, Crop size=448, Stride=224, Rename trick=None2026.0337.3
FreeCPTrain=False, External Model=None, DS Hyperparameter=True, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0337.2
FreeCPExtra Backbone=X, Vision Model Configuration=Single vision model2026.0536.9
DeCLIP2025.0536.4
PnP-OVSSPub. & Year=CVPR’24, Extra data=✗, Extra backbone=BLIP (ViT-L/16), Training free=✓2026.0336.2
CASS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv3 (ViT-B/16), Training free=✓2026.0336
CdamCLIPExtra Backbone=X, Vision Model Configuration=Single vision model2026.0535.2
CLIP-DINOiserTrain=True, External Model=DINO, DS Hyperparameter=False, Image Resize=448, Crop size=448, Stride=224, Rename trick=None2026.0335
ResCLIPTrain=False, External Model=None, DS Hyperparameter=True, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0335
CLIP-DINOiserTrain=true2024.1134.8
CLIP-DINOiser2025.0534.8
CLIP-DINOiserPub. & Year=ECCV’24, Extra data=ImageNet-1K, Extra backbone=DINOv1 (ViT-B/16), Training free=✗2026.0334.8
ResCLIPExtra Backbone=X, Vision Model Configuration=Single vision model2026.0534.7
OVDiffTrain=false, Backbone=ViT-B/162024.1134.6
ClearCLIP + GLATrain=False, External Model=None, DS Hyperparameter=False, Image Resize=448, Crop size=448, Stride=224, Rename trick=None2026.0334.6
LPOSSExtra Backbone=DINO, Vision Model Configuration=Multi vision models2026.0534.3
LaVGPub. & Year=ECCV’24, Extra data=✗, Extra backbone=DINOv1 (ViT-B/8), Training free=✓2026.0334.2
LaVGTrain=False, External Model=None, DS Hyperparameter=False, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0334.2
LPOSS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv1 (ViT-B/16), Training free=✓2026.0333.4
LaVGExtra Backbone=DINO, Vision Model Configuration=Multi vision models2026.0533.3
NACLIPTrain=false, Backbone=ViT-B/162024.1133.2
NACLIP*Pub. & Year=WACV’25, Extra data=✗, Extra backbone=✗, Training free=✓2026.0333.2
NACLIPTrain=False, External Model=None, DS Hyperparameter=False, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0333.2
CASS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv2† (ViT-B/14), Training free=✓2026.0333.1
ClearCLIPExtra Backbone=X, Vision Model Configuration=Single vision model2026.0533
ClearCLIP2025.0533
ClearCLIPPub. & Year=ECCV’24, Extra data=✗, Extra backbone=✗, Training free=✓2026.0333
ClearCLIPTrain=False, External Model=None, DS Hyperparameter=False, Image Resize=448, Crop size=448, Stride=224, Rename trick=None2026.0333
SCLIP + GLATrain=False, External Model=None, DS Hyperparameter=False, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0332.8
CoDeTrain=true2024.1132.3
CoDePub. & Year=CVPR’24, Extra data=CC3M+RedCaps, Extra backbone=✗, Training free=✗2026.0332.3
SFP*Pub. & Year=ICCV’25, Extra data=✗, Extra backbone=✗, Training free=✓2026.0332.1
TCLTrain=True, External Model=None, DS Hyperparameter=False, Image Resize=448, Crop size=448, Stride=224, Rename trick=None2026.0331.6
CLIP-DIYTrain=False, External Model=None, DS Hyperparameter=False, Image Resize=448, Crop size=448, Stride=224, Rename trick=None2026.0331
SCLIPExtra Backbone=X, Vision Model Configuration=Single vision model2026.0530.5
SCLIPTrain=false, Backbone=ViT-B/162024.1130.5
SCLIP2025.0530.5
SCLIP*Pub. & Year=ECCV’24, Extra data=✗, Extra backbone=✗, Training free=✓2026.0330.5
SCLIPTrain=False, External Model=None, DS Hyperparameter=False, Image Resize=336, Crop size=224, Stride=112, Rename trick=With multiple textual prompts per class2026.0330.5
TCLTrain=true2024.1130.4
TCL2025.0530.4
TCLPub. & Year=CVPR’23, Extra data=CC3M+CC12M, Extra backbone=✗, Training free=✗2026.0330.4
GroupViT2025.0527.5
GroupViTPub. & Year=CVPR’22, Extra data=CC12M+RedCaps, Extra backbone=✗, Training free=✗2026.0327.5
GroupViTTrain=True, External Model=None, DS Hyperparameter=False, Image Resize=448, Crop size=448, Stride=224, Rename trick=None2026.0327.5
OVSeg2025.0525.1
SCLIPTrain=false, Backbone=ViT-H/142024.1124.6
FreeDAPub. & Year=CVPR’24, Extra data=COCO Captions, Extra backbone=DINOv2 (ViT-B/14), Training free=✓2026.0324.4
dino.txtExtra Backbone=X, Vision Model Configuration=Single vision model2026.0524.2
MaskCLIPExtra Backbone=X, Vision Model Configuration=Single vision model2026.0520.6
MaskCLIPTrain=false, Backbone=ViT-B/162024.1120.6
MaskCLIP2025.0520.6
MaskCLIPPub. & Year=ECCV’22, Extra data=✗, Extra backbone=✗, Training free=✓2026.0320.6
MaskCLIPTrain=false, Backbone=ViT-H/142024.1116.2
ReCoTrain=false, Backbone=ViT-B/162024.1115.7
ReCo2025.0515.7
ReCoPub. & Year=NeurIPS’22, Extra data=ImageNet-1K, Extra backbone=✗, Training free=✓2026.0315.7
ReCoTrain=True, External Model=None, DS Hyperparameter=False, Image Resize=448, Crop size=448, Stride=224, Rename trick=None2026.0315.7
CaRPub. & Year=CVPR’24, Extra data=✗, Extra backbone=✗, Training free=✓2026.0315.4
CLIP2025.058.1
CLIPPub. & Year=ICML’21, Extra data=✗, Extra backbone=✗, Training free=✓2026.036.5
CLIPTrain=false, Backbone=ViT-B/162024.115.6
CLIPExtra Backbone=X, Vision Model Configuration=Single vision model2026.055.5
OpenCLIPTrain=false, Backbone=ViT-H/142024.115.3