Subhransu S. Bhattacharjee

Results 58 comments of Subhransu S. Bhattacharjee

That actually is good enough.

I think this is something @LiheYoung can help with better. Try an idea called metric point maps, use the (x,y,z) correct to the metric depth that you have and try...

I am not sure, I have never seen such large focal lengths. Try denominations of 10 to figure out what is meaningful to you. If your photo is indeed idk...

The focal lengths are all metric, i do not know how else to answer your question.

The model outputs depth not disparity. I suggest going through the model code once again and understanding the focal lengths and the image size representations.

Were you not using a 3D application? On Wed, 27 Mar, 2024, 10:08 pm karantai, ***@***.***> wrote: > My results were correct from the beginning, but my viewpoint was wrong....

To visualize a point cloud it is imperative that you use a 3D application like Meshlab and correct scales. So no you are wrong, even though your visualization looks more...

Coded it up using the pycolmap library. Will be releasing things very soon. Best Subhransu ________________________________ From: Joshua-Rues ***@***.***> Sent: Friday, August 16, 2024 12:05:46 PM To: colmap/colmap ***@***.***> Cc:...