Reproduce “SpatialVec: A Unified Representation Learning Framework for Geospatial Objects with Boundary-Aware Sampling” 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
Samya Acharja, Kanchan Chowdhury
SpatialVec represents points, lines and polygons for downstream topological relation classification. It samples near boundaries and records signed distance, occupancy and boundary status, while retaining size and location information. On NYC polygon–polygon classification, it reaches 86% accuracy versus Geo2Vec at 70%. The result is one benchmark setting, and ablations do not establish boundary sampling alone caused the gain.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.