UnLearnerMC: Unsupervised learning of dense depth and camera pose using mask and cooperative loss

作者：

Highlights：

• The photometric loop consistency loss is proposed to overcome the moving object interference not included in a pure view synthesis task.

• We combine SegNetMask with the cooperative loss to constrain the moving object area and restrict the number of factors not considered in the mask network.

• UnLearnerMC achieves state-of-the-art results in pose and depth estimation, performing better than previously unsupervised methods.

摘要

•The photometric loop consistency loss is proposed to overcome the moving object interference not included in a pure view synthesis task.•We combine SegNetMask with the cooperative loss to constrain the moving object area and restrict the number of factors not considered in the mask network.•UnLearnerMC achieves state-of-the-art results in pose and depth estimation, performing better than previously unsupervised methods.

论文关键词：Deep learning,Depth estimation,Camera pose,Photometric loop consistency loss,Cooperative loss

论文评审过程：Received 13 February 2019, Revised 31 October 2019, Accepted 6 December 2019, Available online 11 December 2019, Version of Record 24 February 2020.

论文官网地址：https://doi.org/10.1016/j.knosys.2019.105357