Reproduce “CASENet: Deep Category-Aware Semantic Edge Detection” 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
Zhiding Yu, Chen Feng, Ming-Yu Liu, Srikumar Ramalingam
CASENet detects semantic edges by predicting a separate edge map for each category, allowing one boundary to carry multiple labels. It uses shallow features for edge localization and deeper features for category predictions, combining them through shared concatenation and fusion. On Cityscapes validation, it reaches 71.3% mean ODS F-measure versus 68.5% for DSN, but scores are lower for some classes and metrics.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.