Spnet: structure preserving network for depth completion

HIGHLIGHTS

• who: Tao Li and colleagues from the School of Electrical Engineering and Electronic Information, Xihua University, Chengdu, China have published the research work: SPNet: Structure preserving network for depth completion, in the Journal: PLOS ONE of October/25,/2022
• what: To tackle this problem the authors propose a structure preserving network (SPNet) in this paper. Different from them, the authors attempt to capture more useful structural information just from the input RGB images, which contain very rich semantic structures and sharp object boundaries . Specifically, the authors propose a multi-scale gradient extractor (MSGE) to generate . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or E-mail:

  If you don't have an account, you can create one here.