Dense R-MVSNet Details

Author(s):

Yao Yao, Zixin Luo, Shiwei Li, Tianwei Shen, Tian Fang, Long Quan


Description:

A scalable MVS learning pipeline.


Method parameters:

Reconstructed from depth maps from R-MVSNet upsampled by 2.


Used image set:

Reconstructions from provided image set


Publication:

https://arxiv.org/abs/1902.10556


Software or method website:

https://github.com/YoYo000/MVSNet




Error Visualization:

   
   
Precision Recall
   

Change scene:




 

Precision Auditorium




 

Recall Auditorium



© 2017 | tanksandtemples.org