FastFlowNet Android deployment

Have you tried deploying your model on android?

Mar 31 '21 02:03 sieme97

Currently, I have not deployed my model on Android devices. My experience is that correlation operation can only approach real-time when acclerated by parallel computing devices, such as GPU or NPU, etc. This is different from monocular depth networks, which can perform quickly on CPU. I think my future work will consider this application problem.

Mar 31 '21 05:03 ltkong218

Can't we use mobile gpu?

Mar 31 '21 07:03 sieme97

Of course we can use. The Android device with mobile gpu is ideal, and is the trend. The deployment takes some engineering work.

Mar 31 '21 07:03 ltkong218

Thanks for your response. So this means we can't convert the trained models to tflite version directly?

Mar 31 '21 08:03 sieme97

You can load provided pre-trained weights in PyTorch and convert them to the TFLite or ONNX for deployment.

Mar 31 '21 08:03 ltkong218

How to take care of correlation operation when converting the model to tflite?

Apr 07 '21 05:04 sieme97

I have not tried using TFLite. According to my experience, correlation and warping are not standard operations integrated in existing packages. And you should create custom layers to deploy it, which can be referred to documents and examples of TFLite.

Apr 10 '21 03:04 ltkong218

@ltkong218 thanks for your advice. When considering pruned and deployed on mobile phone devices, which model/algorithm would you recommend to use ?

Apr 14 '21 04:04 dragen1860

Oh, I have not consider pruning yet, that may be a future work.

Apr 17 '21 03:04 ltkong218

Hi @sieme97 , @dragen1860, did you manage to run FastFlowNet on Android? I'm also interested in this.

Jan 27 '22 13:01 zdenkas