List your company
ResearchFormerly WizWand

Open Vocabulary Semantic Segmentation on ADE20K A-150

53.99mIoUOracle
11.360422.427733.49544.5623Nov 22, 2024Feb 18, 2025May 18, 2025Aug 15, 2025Nov 12, 2025Feb 9, 2026May 9, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Oracleevaluation protocol=oracle2025.0353.99
dinov3.segVariant=Large, Backbone=dinov3.txt2026.0342.19
ESC-NetVLM=CLIP ViT-L/14, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1141.8
ESCNetVenue=CVPR’25, Variant=Large2026.0341.8
DEDOSVenue=ICCV’25, Variant=Large2026.0339.4
SAM3Venue=ICLR’26, Variant=Large2026.0339
H-CLIPVenue=CVPR’25, Variant=Large2026.0338.4
MAFTP w/ MaskAdapterVLM=CLIP ConvNeXt-L, Training Dataset=COCO-Stuff2024.1238.2
HyperCLIPVenue=CVPR’25, Variant=Large2026.0338.2
SemLAfusion strategy=LoRA merging, tau=0.05, K=5, evaluation protocol=leave-one-out2025.0338.18
CAT-SegVLM=CLIP ViT-L/14, Training Dataset=COCO-Stuff2024.1237.9
CAT-SegVLM=CLIP ViT-L/14, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1137.9
CAT-SegVenue=CVPR’24, Variant=Large2026.0337.9
CAT-Segevaluation protocol=zero-shot2025.0337.83
SemLAfusion strategy=Late Fusion2025.0337.28
Uniform mergingfusion strategy=LoRA merging2025.0337.25
USEVLM=CLIP ViT-L/14, Additional Backbone=DINOv2, SAM, Training Dataset=COCO-Stuff, Additional Dataset=true2024.1137.1
DPSegVenue=CVPR’25, Variant=Large2026.0337.1
OVSNetVenue=ICCV’25, Variant=Large2026.0337.1
MROVSegVLM=CLIP ViT-L/14, Training Dataset=COCO-Stuff2024.1236.9
SMARTVLM=CLIP ViT-L/14, Training Dataset=COCO Panoptic2024.1236.8
FC-CLIP w/ MaskAdapterVLM=CLIP ConvNeXt-L, Training Dataset=COCO Panoptic2024.1236.6
MAFTPVLM=CLIP ConvNeXt-L, Training Dataset=COCO-Stuff2024.1236.3
MAFT+VLM=ConvNeXt-L, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1136.1
MAFT+Venue=ECCV’24, Variant=Large2026.0336.1
Uniformfusion strategy=Late Fusion2025.0336.04
GBAVLM=CLIP ConvNeXt-L, Training Dataset=COCO Panoptic2024.1235.9
MAFTP w/ MaskAdapterVLM=CLIP ConvNeXt-B, Training Dataset=COCO-Stuff2024.1235.6
ESC-NetVLM=CLIP ViT-B/16, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1135.6
SEDVLM=CLIP ConvNeXt-L, Training Dataset=COCO-Stuff2024.1235.3
SEDVLM=ConvNeXt-L, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1135.2
SEDVenue=CVPR’24, Variant=Large2026.0335.2
MAFTPVLM=CLIP ConvNeXt-B, Training Dataset=COCO-Stuff2024.1234.5
FC-CLIPVLM=CLIP ConvNeXt-L, Training Dataset=COCO Panoptic2024.1234.1
FC-CLIPVenue=NeurIPS’23, Variant=Large2026.0334.1
SCANVLM=CLIP ViT-L/14, Additional Backbone=Swin-B, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1133.5
SCANVenue=CVPR’24, Variant=Large2026.0333.5
MAFTVLM=CLIP ViT-L/14, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1133
EBSegVLM=CLIP ViT-L/14, Training Dataset=COCO-Stuff2024.1232.8
EBSegVLM=CLIP ViT-L/14, Additional Backbone=SAM ViT-B, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1132.8
SANVLM=CLIP ViT-L/14, Training Dataset=COCO-Stuff2024.1232.1
SANVLM=CLIP ViT-L/14, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1132.1
SANVenue=CVPR’23, Variant=Large2026.0332.1
EOV-SegVenue=AAAI’25, Variant=Large2026.0332.1
CAT-SegVLM=CLIP ViT-B/16, Training Dataset=COCO-Stuff2024.1231.8
CAT-SegVLM=CLIP ViT-B/16, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1131.8
HyperCLIPVLM=CLIP ViT-B/16, Fine-tuning Space=H2026.0531.7
SEDVLM=CLIP ConvNeXt-B, Training Dataset=COCO-Stuff2024.1231.6
SEDVLM=ConvNeXt-B, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1131.6
SEDVLM=CLIP ConvNeXt-B, Fine-tuning Space=E2026.0531.6
HyRoVLM=CLIP ViT-B/16, Fine-tuning Space=H2026.0531.2
SCANVLM=CLIP ViT-B/16, Additional Backbone=Swin-B, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1130.8
EBSegVLM=CLIP ViT-B/16, Training Dataset=COCO-Stuff2024.1230
EBSegVLM=CLIP ViT-B/16, Additional Backbone=SAM ViT-B, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1130
ODISEVLM=Stable Diffusion, Training Dataset=COCO Panoptic2024.1229.9
ODISEVLM=CLIP ViT-L/14, Additional Backbone=Stable Diffusion, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1129.9
ODISEVenue=CVPR’23, Variant=Large2026.0329.9
OvSegVLM=CLIP ViT-L/14, Training Dataset=COCO-Stuff2024.1229.6
OVSegVLM=CLIP ViT-L/14, Additional Backbone=Swin-B, Training Dataset=COCO-Stuff, Additional Dataset=true2024.1129.6
OVSegVenue=CVPR’23, Variant=Large2026.0329.6
OpenSegVLM=ALIGN, Training Dataset=COCO Panoptic+Loc. Narr.2024.1228.6
SANVLM=CLIP ViT-B/16, Training Dataset=COCO-Stuff2024.1227.5
SANVLM=CLIP ViT-B/16, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1127.5
SANVLM=CLIP ViT-B/16, Additional Backbone=Side Adapter, Fine-tuning Space=E2026.0527.5
FC-CLIPVLM=ConvNeXt-L, Training Dataset=COCO Panoptic, Additional Dataset=false2024.1126.6
OpenSegVLM=ALIGN, Additional Backbone=EfficientNet-B7, Training Dataset=COCO Panoptic, Additional Dataset=true2024.1126.4
OvSegVLM=CLIP ViT-B/16, Training Dataset=COCO-Stuff2024.1224.8
OVSegVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101c, Training Dataset=COCO-Stuff, Additional Dataset=true2024.1124.8
OVSegVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101c, Fine-tuning Space=E2026.0524.8
DeOPVLM=CLIP ViT-B/16, Training Dataset=COCO-Stuff2024.1222.9
DeOPVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101c, Training Dataset=COCO-Stuff-156, Additional Dataset=false2024.1122.9
ZSsegVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1120.5
ZSsegVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101, Fine-tuning Space=E2026.0520.5
ZegFormerVLM=CLIP ViT-B/16, Training Dataset=COCO-Stuff2024.1218
OpenSegVLM=ALIGN, Additional Backbone=ResNet-101, Training Dataset=COCO Panoptic, Additional Dataset=true2024.1117.5
OpenSegVLM=ALIGN, Additional Backbone=ResNet-101, Fine-tuning Space=E2026.0517.5
ZegFormerVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101, Training Dataset=COCO-Stuff-156, Additional Dataset=false2024.1116.9
ZegFormerVLM=CLIP ViT-B/16, Additional Backbone=ResNet-101, Fine-tuning Space=E2026.0516.9
LSeg+VLM=ALIGN, Additional Backbone=ResNet-101, Training Dataset=COCO-Stuff, Additional Dataset=false2024.1113