matrix_city The size of tensor a (1333) must match the size of tensor b (1600) at non-singleton dimension 1.

[hhuzhouhui]

After downsampling in matrix_city, the image resolution is 1600x900. Why is the resolution in sparse/cameras.txt showing as 1920 1080 with the PINHOLE values 2317.6449482429634 2317.6449482429634 960.0 540.0 (1920x1080 resolution)? How can I solve this issue?

RuntimeError: The size of tensor a (1333) must match the size of tensor b (1600) at non-singleton dimension 1.

[hhuzhouhui]

Camera list with one line of data per camera: CAMERA_ID, MODEL, WIDTH, HEIGHT, PARAMS[] Number of cameras: 1 1 PINHOLE 1920.0 1080.0 2317.6449482429634 2317.6449482429634 960.0 540.0

[DekuLiuTesla]

@hhuzhouhui If downsampling is applied, the intrinsics would be automatically adjusted by the code. Perhaps you can check if the code has correctly received correct downsampling rate.

[enhongHan]

I guess when you run estimate depth python, you set the image path to the path with downsampled result. 1333=1920/1.2/1.2. you should run the estimate depth again with origin image path.

[github-actions[bot]]

This issue is stale because it has been open for 30 days with no activity.

[github-actions[bot]]

This issue was closed because it has been inactive for 14 days since being marked as stale.