MiDaS icon indicating copy to clipboard operation
MiDaS copied to clipboard

Code for robust monocular depth estimation described in "Ranftl et. al., Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer, TPAMI 2022"

Results 160 MiDaS issues
Sort by recently updated
recently updated
newest added

I am able to export the MiDaS model w/ DPT large backbone to ONNX using `torch.onnx.export` _almost_ without modifying the model at all. Model constructor looks like: `model = DPTDepthModel(path='dpt_large-midas-2f21e586.pt',...

I'm having trouble figuring out how to define a location to store downloaded models, and load them. Every run it downloads models, and the probability of connections ended by peers,...

I've tried to run MiDaS on Apple silicon with pytorch nightly with mps (metal performance shader) support. I've change two lines of code to get this working: 28 - `...

how to get the output images with JPG extension i am getting it with PNG

i have got midas depth but i need to convert this into distance like meter and centimeter. Can anyone help me?

Fixes #246 by casting numpy.int64 to int.

I get this error when using the model dpt_beit_large_512.pt I use Ubuntu 22.04, Automatic1111 Here is the full report from the console: https://pastebin.com/tF34UqDr

How can i reach this information on midas ?

Hi, I have a question about Imp. (Improvement) calculation. The description says: ![Screenshot 2024-02-05 at 5 59 13 PM](https://github.com/isl-org/MiDaS/assets/15173100/db2f3969-3a2d-47b5-b281-b82540a55269) So, if ɛd is the `zero-shot error` for dataset d, then...

i cant find the pretrained pytroch model