Depth dithering based on texture edge-assisted classification

作者：

Highlights：

• A depth denoising algorithm is proposed for depth maps.

• The depth map is dithered according to its spatial and temporal classification.

• It is efficient in improving the spatial and temporal consistency in depth.

• Exploiting the processed depths, more accurate 3D meshes can be generated.

摘要

•A depth denoising algorithm is proposed for depth maps.•The depth map is dithered according to its spatial and temporal classification.•It is efficient in improving the spatial and temporal consistency in depth.•Exploiting the processed depths, more accurate 3D meshes can be generated.

论文关键词：Depth denoising,Joint spatio-temporal dithering,Spatial classification,Texture edge extraction

论文评审过程：Received 16 June 2015, Revised 7 May 2016, Accepted 7 May 2016, Available online 10 May 2016, Version of Record 6 June 2016.

论文官网地址：https://doi.org/10.1016/j.image.2016.05.006