Open yuqunw opened 2 years ago
Hi, I don't seem to have a copy of that code anymore, sorry. It should be easy to replicate the pointcloud though. The camera poses for each image should be available, and the feature extraction, tracking, and triangulation should be doable with opencv.
I see. I would try to see how I can get a similar preprocessing. Thanks.
Hi, really thank you for sharing your code! I am also interested in the code that you used to generate the sparse point clouds on ScanNet. Could you please also share them? Thanks!