List your company
ResearchFormerly WizWand

Open-vocabulary semantic segmentation on Pascal Context 60

55.4mIoUActiveSAM
5.89618.74831.644.452Mar 23, 2026Apr 6, 2026Apr 20, 2026May 4, 2026May 18, 2026Jun 1, 2026Jun 15, 2026
Updated 4mo ago

Evaluation results

MethodLinks
ActiveSAMVenue=Ours, Backbone (Size)=SAM 3 (PE-L+/14), Training Strategy=Training-free2026.0655.4
SegEarth-OV3Venue=arXiv’25, Backbone (Size)=SAM 3 (PE-L+/14), Training Strategy=Training-free2026.0653.4
SAM 3Venue=ICLR’26, Backbone (Size)=SAM 3 (PE-L+/14), Training Strategy=Training-free2026.0646.1
CorrCLIPVenue=ICCV’25, Backbone (Size)=CLIP + †, Training Strategy=Training-free2026.0644.9
ReMEVenue=ICCV’25, Backbone (Size)=CLIP + §, Training Strategy=Training-free2026.0644.6
TridentVenue=ICCV’25, Backbone (Size)=CLIP + ‡, Training Strategy=Training-free2026.0638.6
RF-CLIPVenue=AAAI’26, Backbone (Size)=CLIP, Training Strategy=Training-free2026.0637.9
Talk2DINOVenue=ICCV’25, Backbone (Size)=CLIP + DINOv2†(ViT-B/14), Training Strategy=Training-based2026.0637.7
SFPVenue=ICCV’25, Backbone (Size)=CLIP, Training Strategy=Training-free2026.0637.2
SPARBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0437.1
CASSVenue=CVPR’25, Backbone (Size)=CLIP + DINO (ViT-B/8), Training Strategy=Training-free2026.0636.7
DIH-CLIPVenue=ICCV’25, Backbone (Size)=CLIP, Training Strategy=Training-free2026.0636
FreeDAVenue=CVPR’24, Backbone (Size)=CLIP + DINOv2 (ViT-B/14), Training Strategy=Training-free2026.0635.3
FreeDAPub. & Year=CVPR’24, Extra data=COCO Captions, Extra backbone=DINOv2 (ViT-B/14), Training free=✓, Mask Refinement=true2026.0335.2
ProxyCLIP*Pub. & Year=ECCV’24, Extra data=✗, Extra backbone=DINOv2† (ViT-B/14), Training free=✓, Mask Refinement=true2026.0335.2
PEARLExtra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=false2026.0335.1
NACLIP*Pub. & Year=WACV’25, Extra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0335
LPOSSVenue=CVPR’25, Backbone (Size)=CLIP + DINO (ViT-B/16), Training Strategy=Training-free2026.0634.6
CASS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv3 (ViT-B/16), Training free=✓, Mask Refinement=true2026.0334.5
LPOSS + SPARBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0434.5
SFP*Pub. & Year=ICCV’25, Extra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0334.3
SPARBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0433.8
ProxyCLIPVenue=ECCV’24, Backbone (Size)=CLIP + DINOv2†(ViT-B/14), Training Strategy=Training-free2026.0633.8
CLIP-DINOiserPub. & Year=ECCV’24, Extra data=ImageNet-1K, Extra backbone=DINOv1 (ViT-B/16), Training free=✗, Mask Refinement=true2026.0333.5
CASS*Pub. & Year=CVPR’25, Extra data=✗, Extra backbone=DINOv2† (ViT-B/14), Training free=✓, Mask Refinement=true2026.0332.6
ClearCLIPVenue=ECCV’24, Backbone (Size)=CLIP, Training Strategy=Training-free2026.0632.6
Pre-trainedBackbone=DINOv3 - ViT-L-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0432.4
CLIP-DINOiserVenue=ECCV’24, Backbone (Size)=CLIP + DINO (ViT-B/16), Training Strategy=Training-based2026.0632.4
NACLIPVenue=WACV’25, Backbone (Size)=CLIP, Training Strategy=Training-free2026.0632.2
Pre-trainedBackbone=DINOv3 - ViT-L-16, Inference=single-pass, Upsampler=bilinear2026.0432.1
Pre-trainedBackbone=DINOv3 - ViT-L-16, Inference=sliding-window, Upsampler=bilinear2026.0431.8
SPARBackbone=DINOv3 - ViT-L-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0431.7
LaVGPub. & Year=ECCV’24, Extra data=✗, Extra backbone=DINOv1 (ViT-B/8), Training free=✓, Mask Refinement=true2026.0331.6
SPARBackbone=DINOv3 - ViT-L-16, Inference=single-pass, Upsampler=bilinear2026.0431.6
LaVGVenue=ECCV’24, Backbone (Size)=CLIP + DINO (ViT-B/8), Training Strategy=Training-free2026.0631.6
SCLIP*Pub. & Year=ECCV’24, Extra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0331.5
LPOSS + Pre-tr.Backbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0430.9
Pre-trainedBackbone=SigLIP2 - ViT-B-16, Inference=sliding-window, Upsampler=bilinear2026.0430.7
CoDeVenue=CVPR’24, Backbone (Size)=CLIP, Training Strategy=Training-based2026.0630.5
TCLPub. & Year=CVPR’23, Extra data=CC3M+CC12M, Extra backbone=✗, Training free=✗, Mask Refinement=true2026.0330.4
SCLIPVenue=ECCV’24, Backbone (Size)=CLIP, Training Strategy=Training-free2026.0630.4
Pre-trainedBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0430.1
CLIPtraseVenue=ECCV’24, Backbone (Size)=CLIP, Training Strategy=Training-free2026.0629.9
SigLIPv2 + UniRefinerViT=So/16, Background category=With background category, Protocol=Zero-shot2026.0529.4
SPARBackbone=OpenCLIP - ViT-B-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0429.2
SAM-CLIPVenue=CVPR’24, Backbone (Size)=CLIP + SAM (ViT-B/16), Training Strategy=Training-based2026.0629.2
SPARBackbone=OpenCLIP - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0428.2
EVA-CLIP + UniRefinerViT=8B/14, Background category=With background category, Protocol=Zero-shot2026.0526.4
Pre-trainedBackbone=OpenCLIP - ViT-B-16, Inference=single-pass, Upsampler=AnyUp [45]2026.0425.3
Pre-trainedBackbone=OpenCLIP - ViT-B-16, Inference=sliding-window, Upsampler=bilinear2026.0425.2
CLIP + UniRefinerViT=G/14, Background category=With background category, Protocol=Zero-shot2026.0525
SegCLIPVenue=ICML’23, Backbone (Size)=CLIP, Training Strategy=Training-based2026.0624.7
Pre-trainedBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0424.5
TCLVenue=CVPR’23, Backbone (Size)=CLIP, Training Strategy=Training-based2026.0624.3
NaFlexBackbone=SigLIP2 - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0423.7
CLIPViT=G/14, Background category=With background category, Protocol=Zero-shot2026.0522.9
MaskCLIPPub. & Year=ECCV’22, Extra data=✗, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0322.6
ReCoPub. & Year=NeurIPS’22, Extra data=ImageNet-1K, Extra backbone=✗, Training free=✓, Mask Refinement=true2026.0321.9
SigLIPv2ViT=So/16, Background category=With background category, Protocol=Zero-shot2026.0521.4
Pre-trainedBackbone=OpenCLIP - ViT-B-16, Inference=single-pass, Upsampler=bilinear2026.0420.3
GroupViTPub. & Year=CVPR’22, Extra data=CC12M+RedCaps, Extra backbone=✗, Training free=✗, Mask Refinement=true2026.0319
GroupViTVenue=CVPR’22, Backbone (Size)=GroupViT (ViT-S/16), Training Strategy=Training-based2026.0618.7
EVA-CLIPViT=8B/14, Background category=With background category, Protocol=Zero-shot2026.0514.1
CaRVenue=CVPR’24, Backbone (Size)=CLIP, Training Strategy=Training-free2026.0613.6
CLIPVenue=ICML’21, Backbone (Size)=CLIP, Training Strategy=Training-free2026.067.8