Estimating layout of cluttered indoor scenes using trajectory-based priors

作者：

Highlights：

• Scene layout estimation using both trajectory data and image features

• No assumptions are considered about the structure of the scene.

• Efficient estimation of the scene structure in the presence of scene clutter

• We show that using line segments we obtain better surface normal.

• Our data and segmentation results will be publicly available for comparison.

摘要

•Scene layout estimation using both trajectory data and image features•No assumptions are considered about the structure of the scene.•Efficient estimation of the scene structure in the presence of scene clutter•We show that using line segments we obtain better surface normal.•Our data and segmentation results will be publicly available for comparison.

论文关键词：Scene segmentation,Trajectory,Scene layout,Semantic context,Conditional random field

论文评审过程：Received 23 September 2013, Revised 2 April 2014, Accepted 21 July 2014, Available online 27 July 2014.

论文官网地址：https://doi.org/10.1016/j.imavis.2014.07.003