WG-FuseNet: Wavelet-Guided and Gated Fusion Network for Road Segmentation

Yu Nie, Jiaqi Sun, Ming Zhu, Yuan Liu, Yuanfu Yuan, Shuhui Jiang, Yan Lu, Jiarong Wang · 2025

Read WG-FuseNet: Wavelet-Guided and Gated Fusion Network for Road Segmentation by Yu Nie, Jiaqi Sun, Ming Zhu, Yuan Liu, Yuanfu Yuan, Shuhui Jiang, Yan Lu, Jiarong Wang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-12788201.txt.

Description

Yu Nie and co-authors present a 2025 computer vision paper on marking roads in images. Fine road edges and surface texture often blur away as networks shrink the image step by step. Fusing color images with depth data by simple stacking also wastes the strengths of each source.

Their network, WG-FuseNet, attacks both problems. A wavelet-based module works in the frequency domain to sharpen edge and texture features across scales. A gated fusion module then blends color and depth features by learned weights rather than fixed sums. The team also tuned the loss function to cut false alarms, such as confusing sidewalks with lanes. Tests in complex scenes showed cleaner road masks with truer boundaries.

Catalog record

CREATORYu Nie, Jiaqi Sun, Ming Zhu, Yuan Liu, Yuanfu Yuan, Shuhui Jiang, Yan Lu, Jiarong Wang
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE54 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC12788201

Browse the full library · About the Tower of Babel