Cyrill Küttel

@Houangnt I'm happy to help. Let me know if you encounter any problems.

There are a couple of options: - To get the highest resolution possible, you could take a picture with the Camera first, then save the image and crop the image...

Hi Did it work? i might have time tomorrow to develop something

It might be possible to maximize the SurfaceView resolution by setting some parameters in Ndk Camera. I actually don't know, I've never tried it. The documentation on Ndk Camera is...

You can load two ncnn models, one for masks, glasses and one for other objects. Then you make inference with both models sequentially (One after the other). But this would...

Unfortunately, I can't solve your problem. I need to see the code.

Seconds this. I would be interested in the steps to take to add a new model.

This is a limitation of [TorchScript supported operators](https://pytorch.org/docs/stable/jit_builtin_functions.html#python-built-in-functions) which are very limited actually. In fact, TorchScript has a lot of limitations, [variable number of arguments](https://github.com/pytorch/pytorch/issues/29637#issue-521347319) is only one of the...

Great! Did you manage to run it on Android / iOS @CoderXXLee ?

does this mean, if we have a different `orig_im_size`, we would have to re-export the model? ```python "orig_im_size": torch.tensor([1500, 2250], dtype=torch.float), ```