Light field occlusion removal network via foreground location and background recovery

作者：

Highlights：

• A light field de-occlusion network was proposed.

• The network breaks the main task into several sub-tasks.

• This network can extract occlusion mask while removing occlusion task.

• Experimental results show that our method achieves superior performances. counterparts.

摘要

•A light field de-occlusion network was proposed.•The network breaks the main task into several sub-tasks.•This network can extract occlusion mask while removing occlusion task.•Experimental results show that our method achieves superior performances. counterparts.

论文关键词：Light field image,Occlusion removal,Foreground location,Background recovery,Deep learning

论文评审过程：Received 30 August 2021, Revised 4 June 2022, Accepted 16 August 2022, Available online 20 August 2022, Version of Record 5 September 2022.

论文官网地址：https://doi.org/10.1016/j.image.2022.116853