3 papers
cs.CV2025
Project-and-Fuse: Improving RGB-D Semantic Segmentation via Graph Convolution Networks
Xiaoyan Jiang, Bohan Wang, Xinlong Wan +3
Most existing RGB-D semantic segmentation methods focus on the feature level fusion, including complex cross-modality and cross-scale fusion modules. However, these methods may cau…
cs.CV2025
FlexiCrackNet: A Flexible Pipeline for Enhanced Crack Segmentation with General Features Transfered from SAM
Xinlong Wan, Xiaoyan Jiang, Guangsheng Luo +2
Automatic crack segmentation is a cornerstone technology for intelligent visual perception modules in road safety maintenance and structural integrity systems. Existing deep learni…
cs.CV2024
Distribution-aware Noisy-label Crack Segmentation
Xiaoyan Jiang, Xinlong Wan, Kaiying Zhu +2
Road crack segmentation is critical for robotic systems tasked with the inspection, maintenance, and monitoring of road infrastructures. Existing deep learning-based methods for cr…