Using existing prior depth map for dense reconstruction?

[Doodle1106]

suppose I have a depth map (near perfect) for each image with metric scale, is there a way to use them directly for dense reconstruction?

[cdcseacave]

yes, if you have also the camera position for each of them

[Doodle1106]

@cdcseacave thank you for your speedy reply! I do have the camera positions for all the images. I would greatly appreciate it if you could kindly guide me on the most suitable starting point for the task. Your expert advice will be invaluable in helping me make progress effectively.

[cdcseacave]

you can use Interface.h file in your project (no instalation required, just copy the file) to export the cameras and depth-maps in the format supported by OpenMVS, and from there just run fusion and mesh reconstruction, etc

you can also use ML, probably slower but could work for you: https://github.com/PRBonn/SHINE_mapping

pls let me know of your progress

[Doodle1106]

@cdcseacave sure, I will keep you updated!

[helloycr]

I have done a lot of work on OpenMVS based on semantic priors. If you need it, please contact me (qq: 956048984)