Multiview Mannequins for Deep Depth Estimation in 360º Videos
Arts & Design
Gaming & Interactive
Research & Education
DescriptionWe estimate depth maps from real-life, monocular, 360º VR videos using a deep neural net architecture trained on 3D point-based renderings of people (called depth “mannequins”) captured with a Multi36 camera setup and processed with a combined pipeline of AI instance segmentation and structure from motion and multi-view stereo methods.