Reproduce “Structure-Aware Network for Lane Marker Extraction with Dynamic Vision Sensor” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Wensheng Cheng, Hao Luo, Wen Yang, Lei Yu, Wei Li
SANet segments lane categories in accumulated images from the DET dynamic vision sensor dataset. It converts pixel-level brightness-change events into 30 ms images and extends SCNN-style message passing with horizontal, vertical, and diagonal propagation. On DET’s held-out test set, SANet reaches 60.86% Mean IoU, 0.42 percentage points above SCNN; this modest gain is limited to accumulated images from one dataset.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.