DSNeRF icon indicating copy to clipboard operation
DSNeRF copied to clipboard
verified publisher icon dunbar12138

Did COLMAP use many images to calculate Sparse 3D Points?

Open blandocs opened this issue 4 years ago • 2 comments

Hi there, thank you for sharing your great work :)

I wonder if you used many (more than two, e.g., 10) images to calculate Sparse 3D points when you had an experiment about a 2-view setting.

When I tried to calculate Sparse 3D points of fern_2v (2-view), the number of detected 3D points is 498, which is quite different from yours directly obtained from "download_example_data.sh". The number of your pre-calculated 3D points is 1081.

image

image

Could you tell me about the experiment setting?

blandocs avatar Aug 25 '21 14:08 blandocs

  • We do not use extra images to obtain 3D points or camera poses.
  • Make sure you are running COLMAP on images of original resolution instead of downscaled ones.
  • We're using the default setting of COLMAP.
  • The result of each COLMAP run varies. You might want to run COLMAP again.

dunbar12138 avatar Aug 26 '21 17:08 dunbar12138

Same issue here. I ran it multiple times on fern_2v using the provided 2 training views and it always produced 498 points. Were you able to get 1k+ points using only these 2 images?

xiyichen avatar Dec 05 '22 09:12 xiyichen