作者: Yoav Y. Schechner , Shree K. Nayar , Peter N. Belhumeur
DOI:
关键词:
摘要: Imaging of objects under variable lighting directions is animportant and frequent practice in computer vision andimage-based rendering. We introduce an approach that significantlyimproves the quality such images. Traditionalmethods for acquiring images illuminationdirections use only a single light source per acquired image.In contrast, our based on multiplexing principle,in which multiple sources illuminate objectsimultaneously from different directions. Thus, objectirradiance much higher. The are thencomputationally demultiplexed. number image acquisitionsis same as single-source method. Theapproach useful imaging dim object areas. givethe optimal code by illumination should be multiplexedto obtain highest output. For n imagescorresponding to sources, noise reduced by\sqrt n/2 relative signal. This reduction translatesto faster acquisition time or increase density illuminationdirection samples. It also enables one lightingwith high directional resolution using practical setups,as we demonstrate experiments.