Open access
Sep 2026
A Lightweight Deep Learning Framework for Parallax-Tolerant Image Stitching
A transformer-based channel attention block improves the discriminative capability of fused features in low-texture regions and enhances global consistency in a lightweight deep stitching framework that integrates multi-scale feature fusion with attention-enhanced matching.
Yi-Liang Wu, Hua-Wang Huang, Zong-Kai Huang et al.
· Electronics · 0 citations