A data-fusion approach to motion-stereo

作者：

Highlights：

• Unconstrained, uncalibrated extension of classical motion-stereo.

• Integrate temporal and spatial information within the same framework.

• Extensive comparison of several disparity confidence measures.

• Cast all depth proxies in a unified framework.

摘要

Highlights•Unconstrained, uncalibrated extension of classical motion-stereo.•Integrate temporal and spatial information within the same framework.•Extensive comparison of several disparity confidence measures.•Cast all depth proxies in a unified framework.

论文关键词：Motion-stereo,Temporal-stereo,Dynamic-stereo,Data fusion,Kalman filter,Parallax

论文评审过程：Received 4 February 2015, Revised 31 January 2016, Accepted 29 February 2016, Available online 3 March 2016, Version of Record 6 April 2016.

论文官网地址：https://doi.org/10.1016/j.image.2016.02.003