List your company
ResearchFormerly WizWand

Open-vocabulary semantic segmentation on Pascal Context-59 (test)

62mIoUCAT-Seg
17.69629.19840.752.202Nov 27, 2023
Updated 8mo ago

Evaluation results

MethodLinks
CAT-SegVLM=ViT-L/14, Feature backbone=Swin-B, Training dataset=COCO-Stuff2023.1162
SEDVLM=ConvNeXt-L, Training dataset=COCO-Stuff2023.1160.6
SANVLM=ViT-L/14, Training dataset=COCO-Stuff2023.1160.2
HIPIEVLM=BERT-B [12], Feature backbone=ViT-H, Training dataset=COCO Panoptic [26]2023.1159.3
FC-CLIPVLM=ConvNeXt-L, Training dataset=COCO Panoptic [26]2023.1158.4
CAT-SegVLM=ViT-B/16, Feature backbone=ResNet-101, Training dataset=COCO-Stuff2023.1157.5
SEDVLM=ConvNeXt-B, Training dataset=COCO-Stuff2023.1157.3
ODISEVLM=ViT-L/14, Training dataset=COCO Panoptic [26]2023.1157.3
OVSegVLM=ViT-L/14, Feature backbone=Swin-B, Training dataset=COCO-Stuff+COCO Caption2023.1155.7
SANVLM=ViT-B/16, Training dataset=COCO-Stuff2023.1153.8
OVSegVLM=ViT-B/16, Feature backbone=ResNet-101c, Training dataset=COCO-Stuff+COCO Caption2023.1153.3
PACLVLM=ViT-B/16, Training dataset=GCC [40]+YFCC [44]2023.1150.1
DeOPVLM=ViT-B/16, Feature backbone=ResNet-101c, Training dataset=COCO-Stuff-1562023.1148.8
SimBaselineVLM=ViT-B/16, Feature backbone=ResNet-101, Training dataset=COCO-Stuff2023.1147.7
Ding et al.VLM=ViT-L/14, Training dataset=COCO Panoptic [26]2023.1145.9
ZegFormerVLM=ViT-B/16, Feature backbone=ResNet-101, Training dataset=COCO-Stuff2023.1145.5
Han et al.VLM=ViT-B/16, Feature backbone=ResNet-101, Training dataset=COCO Panoptic [26]2023.1145.2
OpenSegVLM=ALIGN, Feature backbone=Eff-B7 [43], Training dataset=COCO Panoptic [26]+Loc. Narr. [36]2023.1144.8
ZegFormerVLM=ViT-B/16, Feature backbone=ResNet-101, Training dataset=COCO-Stuff-1562023.1142.8
OpenSegVLM=ALIGN, Feature backbone=ResNet-101, Training dataset=COCO Panoptic [26]+Loc. Narr. [36]2023.1140.1
LSeg+VLM=ALIGN, Feature backbone=ResNet-101, Training dataset=COCO-Stuff2023.1136
GroupViTVLM=ViT-S/16, Training dataset=GCC [40]+YFCC [44]2023.1125.9
SPNetFeature backbone=ResNet-101, Training dataset=PASCAL VOC2023.1124.3
ZS3NetFeature backbone=ResNet-101, Training dataset=PASCAL VOC2023.1119.4