Open wuheng199112068 opened 4 years ago
Hello, this code assumes that you are providing a confidence image as input, and there is no code provided in this repo for computing a confidence image. The best choice for a confidence map likely depends on the application you are targeting. In Section 4 of the Supplement of the paper we describe a reasonable procedure for computing a confidence image from a depth map using a domain transform (Equations 21-23, Figure 1) that often works well, and sometimes just using a confidence map of all 1s works fine.
If put in different images, how do determine the confidence ?Is this the default in you code?