Virtual Reality with Motion Parallax by Optical Flow Depth Generation from Two Spherical Images

Sarthak Pathak, Alessandro Moro, Hiromitsu Fujii, Atsushi Yamashita, Hajime Asama

"Virtual Reality with Motion Parallax by Dense Optical Flow-Based Depth Generation from Two Spherical Images"

2017 IEEE/SICE International Symposium on System Integration

Summary: Video showing results of the proposed motion parallax method.
Two spherical images are captured in a cluttered room environment and the depth map is generated using the proposed method. A perspective video is generated by translation and rotation of the original image based on the depth map generated from optical flow.
The resultant video shows accurate and natural-looking motion parallax, which is a necessary depth perception cue for virtual reality systems.