Jul 2026
Self-visual prompting for training-free recursive weakly supervised semantic segmentation
Self-visual prompting (SVP), a training-free CAM generation paradigm that employs a recursive self-bootstrapping strategy to shift optimization from model parameters to the input context, and progressively refines the quality of CAMs.
Congwei Zhang, Zhibin Quan, Yuncon Yao et al.
· Multimedia Systems · 0 citations