A Unified Spatial-Angular Structured Light for Single-View Acquisition of Shape and Reflectance
Xianmin Xu, Yuxin Lin, Haoyang Zhou, Chong Zeng, Yaxin Yu, Kun Zhou and Hongzhi Wu
CVPR 2023.
Patent Pending.
Abstract
We propose a unified structured light, consisting of an LED array and an LCD mask, for high-quality acquisition of both shape and reflectance from a single view. For geometry, one LED projects a set of learned mask patterns to accurately encode spatial information; the decoded results from multiple LEDs are then aggregated to produce a final depth map. For appearance, learned light patterns are cast through a transparent mask to efficiently probe angularly- varying reflectance. Per-point BRDF parameters are differentiably optimized with respect to corresponding measurements, and stored in texture maps as the final reflectance. We establish a differentiable pipeline for the joint capture to automatically optimize both the mask and light patterns to- wards optimal acquisition quality. The effectiveness of our light is demonstrated with a wide variety of physical objects. Our results compare favorably with state-of-the-art techniques.
Downloads
Paper [.PDF, 7.9MB]
Supplemental Material [.PDF, 0.8MB]
Video [Youtube/Bilibili]
Bibtex [.BIB]
Xianmin Xu, Yuxin Lin, Haoyang Zhou, Chong Zeng, Yaxin Yu, Kun Zhou and Hongzhi Wu
CVPR 2023.
Patent Pending.
Abstract
We propose a unified structured light, consisting of an LED array and an LCD mask, for high-quality acquisition of both shape and reflectance from a single view. For geometry, one LED projects a set of learned mask patterns to accurately encode spatial information; the decoded results from multiple LEDs are then aggregated to produce a final depth map. For appearance, learned light patterns are cast through a transparent mask to efficiently probe angularly- varying reflectance. Per-point BRDF parameters are differentiably optimized with respect to corresponding measurements, and stored in texture maps as the final reflectance. We establish a differentiable pipeline for the joint capture to automatically optimize both the mask and light patterns to- wards optimal acquisition quality. The effectiveness of our light is demonstrated with a wide variety of physical objects. Our results compare favorably with state-of-the-art techniques.
Downloads
Paper [.PDF, 7.9MB]
Supplemental Material [.PDF, 0.8MB]
Video [Youtube/Bilibili]
Bibtex [.BIB]