SOURCE-LINKED INTELLIGENCE

Lightweight Machine Learning-Driven Monocular Sidewalk Path Extraction for Embedded Micromobility Navigation

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

Sidewalk-scale path extraction demands perception and planning that run reliably on compact, low-power hardware in cluttered, map-sparse environments. We present a monocular vision pipeline for sidewalk path extraction in micromobility systems that progresses through three design iterations, from a skeleton-graph baseline through distance-transform corridor planning to a lightweight image-space architecture, and provides a systematic comparison of five path-planning methods across both bird's-eye-view (BEV) and image-space domains. A compact SegFormer-B0 student model, trained with a semi-supe

Read original source ↗ Open in workspace

recordType

paper

region

Global