search

nianticlabs / manydepth

[CVPR 2021] Self-supervised depth estimation from short sequences
Other
626 stars 84 forks source link

Depth evaluation for single frame mode? #51

Closed LeeBY68 closed 2 years ago

LeeBY68 commented 2 years ago

Thanks for the wonderful work!

I have a question for the depth evaluation: When I evaluate the depth performance of a single image, which has no previous frame, I set "--zero_cost_vulome" and "--num_matching_frames = 0" for the evaluation options. However, the "evaluate_depth" encouter a failure: because "frames_to_load[1:]" is empty, the "lookup_frames" receive the empty Tensor-list.

What should I set or change for the single frame mode evaluation, where the test frame has completely no previous frames or future frames?

daniyar-niantic commented 2 years ago

Hi. You should use test_simple.py. You can specify the target image for source_image_path (as well as target_image_path) but use mode='mono' so that the source image is ignored during prediction.

zshn25 commented 2 years ago

Does the —static_camera flag simulate single frame inference during evaluation? I would like to run the evaluate_depth.py but with single frame as input