HSCM-Lane: A ResNet-based encoder-decoder architecture with multi-level shifted-window context modeling for pixel-wise lane segmentation
Quang Toai Ton, Thanh Hien Vu, Thanh Nguyen Vu, Tuong Le · 2026
Read HSCM-Lane: A ResNet-based encoder-decoder architecture with multi-level shifted-window context modeling for pixel-wise lane segmentation by Quang Toai Ton, Thanh Hien Vu, Thanh Nguyen Vu, Tuong Le (2026) 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/plos-pone-0359959.txt.
Description
Quang Toai Ton, Thanh Hien Vu, Thanh Nguyen Vu, and Tuong Le published this research article in PLOS ONE in 2026. It presents HSCM-Lane, a computer vision system for lane segmentation, marking exactly which pixels in a road image belong to lane lines. The system is built on a ResNet encoder-decoder design.
Added to that base is context modeling that uses shifted windows at several levels, a way for the network to consider both fine line detail and the wider road scene. Lane marking detection is a core skill for driver assistance and self-driving cars. Engineers comparing segmentation architectures can judge whether this context modeling earns its extra complexity on road images.
Catalog record
| CREATOR | Quang Toai Ton, Thanh Hien Vu, Thanh Nguyen Vu, Tuong Le |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 97141 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |