CVPR2020

Adversarial Texture Optimization From RGB-D Scans

Jingwei Huang, Justus Thies, Angela Dai, Abhijit Kundu, Chiyu Max Jiang, Leonidas J. Guibas, Matthias Nießner, Thomas A. Funkhouser

DOI 出版方

摘要

Ours Zhou and Koltun Figure 1. Our goal is to reconstruct high-quality textures from an RGB-D scan. Unlike traditional methods which optimize for a parametric color map to reduce misalignment error (Zhou and Koltun [36]), we learn a misalignment-tolerant discriminator, producing sharper textures.