sacmehta / espnetv2-coreml

Semantic segmentation on iPhone using ESPNetv2

Real-time semantic segmentation using ESPNetv2 on iPhone

This repository provides a real-time demo of ESPNetv2 on iPhone (tested only on iPhone7). Below are some illustrations.

Real-time semantic segmentation using ESPNetv2 on iPhone7

Model details

The COREML ESPNetv2 model takes an RGB image of size 256x256 as an input and produces an output of size 256x256 in real-tim. The model learns about 0.79 million parameters and performs roughly 337 million FLOPs to generate the segmentation mask. The model is trained using PyTorch on the PASCAL VOC 2012 dataset and achieves a segmentation score of 63.36, which is measured in terms of mean interesection over union (mIOU).

Several pre-trained models are provided in our EdgeNets repository.

Contributions

If you are familiar with iOS application development and wants to improve the design or contribute in some way, please do so by creating a pull request. We welcome contributions.

License

The code and models are released under the same license as EdgeNets.

Acknowledgement

Many thanks to Srini and Hanna for their support and help as always.

espnetv2-coreml's Issues

inference real time?

I run the demo on iphone 6s, I counted the inferencing time of the model, it was about 120ms, Is this time correct?
let startTime = CFAbsoluteTimeGetCurrent()

try imageRequestHandler.perform(self.requests)

let endTime = CFAbsoluteTimeGetCurrent()
debugPrint("inference time：%f ms", (endTime - startTime)*1000)

Hello, request for previous version of the app.

Hello, I see that in the readme file there is a demo previewed in illustrations. It is a little different than the repository inside. Do you have a version when recognized mask is covered above the image like it is on your illustrations?;)

I would like to see your example as it presented in read me;) Could you share that version of repo with me?

Thank you very much.