Dehazing Cost Volume for Deep Multi-view Stereo in Scattering Media

Yuki Fujimura, Motoharu Sonogashira, Masaaki Iiyama; Proceedings of the Asian Conference on Computer Vision (ACCV), 2020

We propose a learning-based multi-view stereo (MVS) method in scattering media such as fog or smoke with a novel cost volume, called the dehazing cost volume. An image captured in scattering media degrades due to light scattering and attenuation caused by suspended particles. This degradation depends on scene depth; thus it is difficult for MVS to evaluate photometric consistency because the depth is unknown before three-dimensional reconstruction. Our dehazing cost volume can solve this chicken-and-egg problem of depth and scattering estimation by computing the scattering effect using swept planes in the cost volume. Experimental results on synthesized hazy images indicate the effectiveness of our dehazing cost volume against the ordinary cost volume regarding scattering media. We also demonstrated the applicability of our dehazing cost volume to real foggy scenes.