List your company
ResearchFormerly WizWand

Open-Vocabulary Semantic Segmentation on COCOStuff (val)

46.4mIoUCONCEPTBANK
2.7214.0625.436.74Oct 2, 2024Dec 23, 2024Mar 15, 2025Jun 5, 2025Aug 26, 2025Nov 16, 2025Feb 6, 2026
Updated 8mo ago

Evaluation results

MethodLinks
CONCEPTBANKBackbone (Size)=SAM3 (PE-L+/14), Optimization protocol=Parameter-free2026.0246.4
SegEarth-OV3Size=PE-L+/142025.1242.8
CorrCLIPBackbone (Size)=CLIP (ViT-L/14) + DINO (ViT-B/8) + SAM2 (Hiera-L), Optimization protocol=Parameter-free2026.0234
CorrCLIPSize=ViT-L/14, Training Protocol=Training-free2025.1234
ReMEBackbone (Size)=CLIP + DINOv2 (ViT-L/14) + SAM (ViT-L/16), Optimization protocol=Parameter-free2026.0233.3
SAM3Backbone (Size)=SAM3 (PE-L+/14), Optimization protocol=Parameter-free2026.0233.3
CorrCLIPSize=ViT-H/14, Training Protocol=Training-free2025.1232.7
CorrCLIPSize=ViT-B/16, Training Protocol=Training-free2025.1231.6
SAM-CLIPBackbone (Size)=CLIP + SAM (ViT-B/16), Optimization protocol=Training-based2026.0231.5
Talk2DINOBackbone (Size)=CLIP + DINOv2 with registers (ViT-B/14), Optimization protocol=Training-based2026.0230.2
FreeDASize=ViT-L/14, Training Protocol=Training-free2025.1228.8
CLIPerSize=ViT-L/14, Training Protocol=Training-free2025.1228.7
TridentSize=ViT-H/14, Training Protocol=Training-free2025.1228.6
TridentBackbone (Size)=CLIP + DINO (ViT-B/16) + SAM (ViT-B/16), Optimization protocol=Parameter-free2026.0228.3
TridentSize=ViT-B/16, Training Protocol=Training-free2025.1228.3
RF-CLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.0227.5
CLIPerSize=ViT-B/16, Training Protocol=Training-free2025.1227.5
SC-CLIPSize=ViT-L/14, Training Protocol=Training-free2025.1226.9
ProxyCLIPSize=ViT-H/14, Training Protocol=Training-free2025.1226.8
CASSBackbone (Size)=CLIP + DINO (ViT-B/8), Optimization protocol=Parameter-free2026.0226.7
DIH-CLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.0226.7
CASSSize=ViT-B/16, Training Protocol=Training-free2025.1226.7
SC-CLIPSize=ViT-B/16, Training Protocol=Training-free2025.1226.6
ProxyCLIPSize=ViT-B/16, Training Protocol=Training-free2025.1226.5
SFPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.0226.4
LPOSSBackbone (Size)=CLIP + DINO (ViT-B/16), Optimization protocol=Parameter-free2026.0225.9
FreeDABackbone (Size)=CLIP + DINOv2 (ViT-B/14), Optimization protocol=Parameter-free2026.0225.7
NACLIPSize=ViT-B/16, Training Protocol=Training-free2025.1225.7
ProxyCLIPSize=ViT-L/14, Training Protocol=Training-free2025.1225.6
ProxyCLIPBackbone (Size)=CLIP + DINOv2 with registers (ViT-B/14), Optimization protocol=Parameter-free2026.0225.4
ClearCLIPUpsampling=SimFeatUp2024.1025.1
ResCLIPSize=ViT-B/16, Training Protocol=Training-free2025.1224.7
CLIP-DINOiserBackbone (Size)=CLIP + DINO (ViT-B/16), Optimization protocol=Training-based2026.0224.6
CLIP-DINOiserSize=ViT-B/16, Training Protocol=Training-based2025.1224.6
CLIPtraseSize=ViT-B/16, Training Protocol=Training-free2025.1224.1
ClearCLIPUpsampling=Original2024.1023.9
CoDeBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Training-based2026.0223.9
ClearCLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.0223.9
CoDeSize=ViT-B/16, Training Protocol=Training-based2025.1223.9
ClearCLIPSize=ViT-B/16, Training Protocol=Training-free2025.1223.9
ResCLIPSize=ViT-L/14, Training Protocol=Training-free2025.1223.4
NACLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.0223.3
LaVGBackbone (Size)=CLIP + DINO (ViT-B/8), Optimization protocol=Parameter-free2026.0223.2
LaVGSize=ViT-B/16, Training Protocol=Training-free2025.1223.2
CLIPtraseBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.0222.8
SCLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.0222.4
TCLSize=ViT-B/16, Training Protocol=Training-based2025.1222.4
SCLIPSize=ViT-B/16, Training Protocol=Training-free2025.1222.4
SCLIPUpsampling=SimFeatUp2024.1022
SCLIPUpsampling=Original2024.1021.1
TCLUpsampling=Original2024.1019.6
TCLBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Training-based2026.0219.6
MaskCLIPUpsampling=SimFeatUp2024.1018
MaskCLIPUpsampling=Original2024.1016.4
MaskCLIPSize=ViT-B/16, Training Protocol=Training-free2025.1216.4
GEMBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.0215.7
GroupViTBackbone (Size)=GroupViT (ViT-S/16), Optimization protocol=Training-based2026.0215.3
RecoUpsampling=Original2024.1014.8
CLIPBackbone (Size)=CLIP (ViT-B/16), Optimization protocol=Parameter-free2026.027.2
CLIPSize=ViT-B/16, Training Protocol=Training-free2025.124.4