Eljas Hyyrynen

Your upper image shows a warning, not an error. A warning is just a warning. Could you try running `eval.py` with pytorch model instead of onnx model? It's not meant...

Could you please elaborate what you have tried - Your system specs - The exact commands you used for installation - The exact commands you used for running the code...

Maybe the NN is missing such training samples. It's not perfect in all imaginable conditions. Maybe you have found a "weakness" in the model. Could you create a dataset of...

You can check the .onnx model graph by using this awesome tool: https://netron.app/

@keishatsai what are your goals? Do you want to achieve a higher accuracy? You can check the annotation from the model they used for training. You can also check the...

I, too, have to problem with a powerful PC, Ubuntu 20.04, tf 1.14, CUDA 11. It sometimes gets a weird lag peak even though mostly it's running with nice stable...

@BhuvaneshwariBhaskaran which python version are you using? What is your output for `pip freeze`?

@abnerdesigner1992 are you talking about ELG or DPG paper? I think they used the simulator to train the model entirely. The MPIIGaze training data can be downloaded using the script...

@l35633 could you share the code you used to try to save the model and then read the model? Did funfuzzes suggestion help you? I got the same error. Google...

If there are many faces how do you intend on choosing the face? Always the same face than the first detected one or the face with best score? For webcam...