Show HN: My compact, bio inspired distributed edge detection algorithm

Category: ai-ml

Tags: edge-detection, computer-vision, pytorch, bio-inspired, image-processing

Score: 5.3/10 (Innovation: 5, Technical: 6, Documentation: 6, Utility: 4)

HaCI is a bio-inspired, four-stage edge detection pipeline that uses harmonic contrast, oriented pooling, Naka-Rushton gain control, and ridge back-projection to produce boundary maps. It bundles a pretrained model and reports competitive ODS on BRIND/BSDS-style benchmarks, making it a self-contained research artifact.

Target audience: computer vision researchers, ML engineers, image processing developers

Repository: https://github.com/RenderBear/haci · Python · Apache-2.0 · 1 stars

View on Hacker News