Reproduce “On Detection of Faint Edges in Noisy Images” 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

Nati Ofir, Meirav Galun, Sharon Alpert, Achi Brandt, Boaz Nadler, Ronen Basri

The authors present an edge detector that searches candidate paths using average intensity differences on either side, accumulating evidence along noisy boundaries. Its threshold decreases with length for straight edges but remains positive asymptotically for exponentially large curved-edge searches. In 129 × 129 simulations, the stringent curved method achieved average F-measure 0.73, versus 0.57 for Canny, on the tested patterns.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.