YogAI is a virtual yoga instructor on a raspberry pi smart mirror. Using an Openpose tensorflow implementation forked from ildoonet/tf-pose-estimation, we can guide and instruct a student during their yoga session and improve their form.
We've have a tflite implementation! For much faster inference, please see this repo
You need dependencies below.
- python3
- tensorflow 1.4.1+
- opencv3, protobuf, python3-tk
- raspberry pi 3
- any webcam of choice
- a speaker with aux cord
- computer screen
- one way mirror + frame (optional)
$ git clone https://www.github.com/ildoonet/YogAI
$ cd YogAI
$ pip3 install -r requirements.txtCMU's model graphs are too large for git, so I uploaded them on an external cloud. You should download them if you want to use cmu's original model. Download scripts are provided in the model folder.
$ cd models/graph/cmu
$ bash download.sh
The Hackster post will show you how to obtain training samples for your desired poses. Use the yoga_pose_data.py script to transform the images into Posenet point arrays with labels. The YogAI_knn.ipynb is a jupyter notebook to help you train a KNN to classify yoga poses.
See : etcs/training.md
[1] https://github.com/CMU-Perceptual-Computing-Lab/openpose
[2] Training Codes : https://github.com/ZheC/Realtime_Multi-Person_Pose_Estimation
[3] Custom Caffe by Openpose : https://github.com/CMU-Perceptual-Computing-Lab/caffe_train
[4] Keras Openpose : https://github.com/michalfaber/keras_Realtime_Multi-Person_Pose_Estimation
[1] Arxiv Paper : https://arxiv.org/abs/1701.00295
[2] https://github.com/DenisTome/Lifting-from-the-Deep-release
[1] Original Paper : https://arxiv.org/abs/1704.04861
[2] Pretrained model : https://github.com/tensorflow/models/blob/master/slim/nets/mobilenet_v1.md
[1] Tensorpack : https://github.com/ppwwyyxx/tensorpack
[1] Freeze graph : https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/tools/freeze_graph.py
[2] Optimize graph : https://codelabs.developers.google.com/codelabs/tensorflow-for-poets-2