HyperAIHyperAI
AccueilActualitésArticles de recherche récentsTutorielsEnsembles de donnéesWikiSOTAModèles LLMClassement GPUÉvénements
Recherche
À propos
HyperAIHyperAI

Unsupervised Semantic Segmentation With 9

Métriques

mIoU

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle
mIoU
Paper TitleRepository
MaskCLIP16.4Extract Free Dense Labels from CLIP-
TCL22.4Learning to Generate Text-grounded Mask for Open-world Semantic Segmentation from Only Image-Text Pairs-
TTD (MaskCLIP)19.4TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias-
COSMOS ViT-B/1623.2COSMOS: Cross-Modality Self-Distillation for Vision Language Pre-training-
TagAlign25.3TagAlign: Improving Vision-Language Alignment with Multi-Tag Classification-
ReCo14.8ReCo: Retrieve and Co-segment for Zero-shot Transfer-
Trident28.6Harnessing Vision Foundation Models for High-Performance, Training-Free Open Vocabulary Segmentation-
TTD (TCL)23.7TTD: Text-Tag Self-Distillation Enhancing Image-Text Alignment in CLIP to Alleviate Single Tag Bias-
ProxyCLIP26.8ProxyCLIP: Proxy Attention Improves CLIP for Open-Vocabulary Segmentation-
GroupViT11.1GroupViT: Semantic Segmentation Emerges from Text Supervision-
0 of 10 row(s) selected.
Unsupervised Semantic Segmentation With 9 | SOTA | HyperAI