How to extract vidit_depth maps?

[lrxsxdl]

Thanks for your excellent work! Could you please provide some instructions about how to extract a vidit_depth map of an image?

I have tried the method from [https://huggingface.co/vinvino02/glpn-nyu], which is used to extract the vivit_depth map in the dataset [https://huggingface.co/datasets/Nahrawy/VIDIT-Depth-ControlNet] you utilized for finetuing. But I got the depth map with a shape of [H, W], which differs from the need map with a shape of [H,W,3].