Reproduce “NEF: Neural Edge Fields for 3D Parametric Curve Reconstruction from Multi-view 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
Yunfan Ye, Renjiao Yi, Zhirui Gao, Chenyang Zhu, Zhiping Cai, Kai Xu
NEF reconstructs 3D shape curves from calibrated images by learning an edge field whose rendered edges match 2D edge maps, avoiding explicit cross-view edge matching and 3D edge labels. It converts recovered edge points into parametric curves. On synthetic ABC-NEF, NEF reports an F-score of 0.9044 versus DEF’s 0.8009, though methods use different input modalities.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.