Reproduce “GPU-Native Real-Time Textured Mesh Reconstruction with Seam-Aware Multi-View Radiometric Correction” 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
Hiroyuki Sato
The authors present a GPU-resident pipeline for producing textured meshes from simultaneous RGB-D views each frame. A compact adjacency structure supports GPU mesh processing, while a spatial gain field estimates relative camera brightness differences at seams. With five Kinect inputs and 0.02 m voxels, processing takes 13.1 ms per frame, below the 33 ms budget; at 0.005 m, it takes 138.9 ms and is not real-time.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.