List your company
ResearchFormerly WizWand

Open-vocabulary Semantic Segmentation on Pascal VOC 21

83.2mIoUFG-CLIP 2
24.9640.0855.270.32Oct 13, 2025Nov 10, 2025Dec 9, 2025Jan 6, 2026Feb 4, 2026Mar 4, 2026Apr 2, 2026
Updated 5mo ago

Evaluation results

MethodLinks
FG-CLIP 2Backbone=ViT-So/162025.1083.2
SigLIP 2Backbone=ViT-L/162025.1082.3
CLIPBackbone=ViT-L/142025.1081.8
FG-CLIPBackbone=ViT-L/142025.1081.8
FG-CLIP 2Backbone=ViT-L/162025.1081.8
FG-CLIP 2Backbone=ViT-B/162025.1081.1
CLIPBackbone=ViT-B/162025.1078.3
FG-CLIPBackbone=ViT-B/162025.1077.9
SigLIP 2Backbone=ViT-B/162025.1075.8
CLIP-DINOiserPub. & Year=ECCV’24, Extra data=ImageNet-1K, Extra backbone=DINOv1 (ViT-B/16), Training free=✗, Mask Refinement=true2026.0364.6
NACLIP*Pub. & Year=WACV’25, Extra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0364.1
PEARLExtra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=false2026.0364.1
CASS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv3 (ViT-B/16), Training free=✓, Mask Refinement=true2026.0362.5
LaVGPub. & Year=ECCV’24, Extra data=✗, Extra backbone=DINOv1 (ViT-B/8), Training free=✓, Mask Refinement=true2026.0362.1
ProxyCLIP*Pub. & Year=ECCV’24, Extra data=✗, Extra backbone=DINOv2† (ViT-B/14), Training free=✓, Mask Refinement=true2026.0362
SCLIP*Pub. & Year=ECCV’24, Extra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0361.7
SFP*Pub. & Year=ICCV’25, Extra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0358.8
CASS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv2† (ViT-B/14), Training free=✓, Mask Refinement=true2026.0358.7
TCLPub. & Year=CVPR’23, Extra data=CC3M+CC12M, Extra backbone=✗, Training free=✗, Mask Refinement=true2026.0355
FreeDAPub. & Year=CVPR’24, Extra data=COCO Captions, Extra backbone=DINOv2 (ViT-B/14), Training free=✓, Mask Refinement=true2026.0352
LPOSS + SPARBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0451.2
GroupViTPub. & Year=CVPR’22, Extra data=CC12M+RedCaps, Extra backbone=✗, Training free=✗, Mask Refinement=true2026.0351.1
SPARBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0451
SPARBackbone=OpenCLIP - ViT-B-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0450.3
Pre-trainedBackbone=OpenCLIP - ViT-B-16, Inference=sliding-window, Upsampler=bilinear2026.0448.5
SPARBackbone=OpenCLIP - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0448.5
Pre-trainedBackbone=OpenCLIP - ViT-B-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0448.5
SPARBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0447.3
Pre-trainedBackbone=DINOv3 - ViT-L-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0446.2
LPOSS + Pre-tr.Backbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0446.1
Pre-trainedBackbone=DINOv3 - ViT-L-16, Inference=single-pass, Upsampler=bilinear2026.0446
Pre-trainedBackbone=SigLIP2 - ViT-B-16, Inference=sliding-window, Upsampler=bilinear2026.0445
Pre-trainedBackbone=OpenCLIP - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0443.1
SPARBackbone=DINOv3 - ViT-L-16, Inference=single-pass, Upsampler=bilinear2026.0443.1
SPARBackbone=DINOv3 - ViT-L-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0442.8
Pre-trainedBackbone=DINOv3 - ViT-L-16, Inference=sliding-window, Upsampler=bilinear2026.0442.6
Pre-trainedBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0442.4
MaskCLIPPub. & Year=ECCV’22, Extra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0337.2
Pre-trainedBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0436.1
NaFlexBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0435.8
ReCoPub. & Year=NeurIPS’22, Extra data=ImageNet-1K, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0327.2