Lifting-Posture-Prediction

Baseline NN/cGAN/cVAE

Prerequisite

Python 3
Torch

Usage

Download weights to folder 'weights'.
For ubuntu or wins terminal:

Clone this repo:
git clone [email protected]:LLDavid/Lifting-Posture-Prediction.git
Run:
python test.py --type "baseline" --HHW 1.70 0.45 0.45 0.45 0.45

For python IDE on wins:

Download this repo.
Open 'test.py'. Edit the input argument 'type' and 'HHW' as you want:
model_test(type="baseline", HHW=[1.67, 0.6,0.6,0.45,0.45])
Then run the script.

Arguments

'type' has 3 options: "baseline", "cVAE", "cGAN"
'HHW' denote [subject height, left hand height, right hand height, left hand width, right hand width]
For more details, please refer to our paper (ongoing).

Example.

Calculating joint angles and segment lengths from bony landmarks. (Link)

Selected Hypaer-parameters and Candidates.