Zhenyu Li
Zhenyu Li
That sucks. Please first ensure that the value of your depth maps lies in 0-10 after dividing it by the `depth_scale=1000`. Then, try to set `scale_up=True` presented in issue#20 to...
1. You can refer to my new repo about LiteDepth. That's an example of using this toolbox and adopting a new dataset. 2. The simplest way in my view is...
Actually, I just follow previous works, mainly adopting RGB-D camera or point cloud projections. I think a pre-trained big model can also be OK and you can also refer to...
Get it. It seems I forget to include the config. Upload later.
I'm now revising this paper. Thanks for your questions. I may improve my description of the 'auxiliary classification' in this version. Let me try to answer you more directly. Considering...
Close it for now.
Refer to [1](https://github.com/zhyever/LiteDepth/blob/958cf4ea12060e80eacf08a0f4077003db9d9d11/projects/toolbox_plugin/datasets/mobile_ai_2022_dataset.py#L21), where we first save the images with np.int16. (.npy) Then, refer to [2](https://github.com/zhyever/LiteDepth/blob/958cf4ea12060e80eacf08a0f4077003db9d9d11/tools/convert_format.py#L31), where we actually load the result(.npy) again and use cv2 to save it as...
Close it for now.
Check here: https://github.com/zhyever/Monocular-Depth-Estimation-Toolbox/tree/main/splits I provide them in the `split` folder.
Hope it is helpful. Close this issue for now.