[3DSAp2/3Dp2] 3D and Hyper-realistic Displays and Applications 2

[3DSAp2/3Dp2-2] Unsupervised Monocular Depth Estimation for Autonomous Driving

Chih-Shuan Huang¹, *Wan-Nung Tsung¹, Wei-Jong Yang¹, Chin-Hsing Chen¹ (1. National Cheng Kung University (Taiwan))

キーワード：Autonomous Driving, Depth Estimation, Disparity, 3D image

3D technology with range information has become a staple requirement in computer vision. For this reason, we believe that the depth information can effectively improve the vision capabilities for many applications. In this paper, we proposed an unsupervised monocular depth estimation network to extract the depth map of street views.