Unsupervised Semantic Segmentation With 12

mIoU

Results

Performance results of various models on this benchmark

		Paper Title	Code
Trident	40.1	Harnessing Vision Foundation Models for High-Performance, Training-Free Open Vocabulary Segmentation
ProxyCLIP	35.4	ProxyCLIP: Proxy Attention Improves CLIP for Open-Vocabulary Segmentation

0 of 2 row(s) selected.

mIoU

Performance results of various models on this benchmark

		Paper Title	Code
Trident	40.1	Harnessing Vision Foundation Models for High-Performance, Training-Free Open Vocabulary Segmentation
ProxyCLIP	35.4	ProxyCLIP: Proxy Attention Improves CLIP for Open-Vocabulary Segmentation

0 of 2 row(s) selected.