tensorflow-yolo-ios icon indicating copy to clipboard operation
tensorflow-yolo-ios copied to clipboard

Real time object detection demo App with Yolo on iOS based on tensorflow framework

Slack Channel for Deep Learning Communication:

https://deep-learning-geeks-slack.herokuapp.com/

Tensorflow-Yolo-iOS

Demo

Dependencies

Tensorflow 1.1.0

Swift 3

DeepBelief

Git LFS

Note

  1. This repo is build and modifided based on enVision, which is built with tensorflow < 1.* ;

  2. Since, there's a quite big dependencies folder with tensorflow v1.1.0 static built and re-organization, we used Git LFS to store all the big files. It means, when you try to build this project locally, make sure you have git lfs installed and fetch all codes ;

Do following:

brew install git-lfs
git lfs install
git lfs clone https://github.com/KleinYuan/tensorflow-yolo-ios
  1. A frozen tiny-yolo tensorflow model is provided by default, with VOC data trained. If you want to do an end-to-end (meaning, train yolo on darknet -> translate -> freeze model -> implement on iOS), you may need this