1
Return

CTP-CLIP: A clustered text prompt framework for weakly supervised semantic segmentation

delete2026-08-05
delete0
PRE
AI
X
Xuefei Bai
X
Xinkai Huo
W
Wenjie Xu
W
Wenjian Wang *
DOI：10.1016/j.neucom.2026.134729delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• A clustered prompt framework is proposed for CLIP-based WSSS. • Hierarchical text priors improve cross-modal localization. • Stage-wise attention propagation refines CAMs and pseudo-labels. • Adaptive multi-level fusion better exploits frozen CLIP features. • Competitive results are achieved on PASCAL VOC 2012 and MS COCO 2014.

Journal

Neurocomputing cover
Neurocomputing
IF:
6.5
Papers:
2.5W
Citations:
6.5W

Organization

S
Shanxi Province
Scholars:
16
Papers: 8
Citations: 0
S
Shanxi University
Scholars:
1.2W
Papers: 8.2K
Citations: 1.2W
Cited Papers

Cited Papers

Citing Papers

Citing Papers