MotionPrediction

Reference project

Setup

• OS : Ubuntu 18.04 or 20.04

Gym

1. Create python virtual environment with python 3.8
2. Install pytorch
3. Install Isaac Gym

   cd issacgym/python
   pip install -e .

4. Install Issac Gym Envs

   git clone https://github.com/NVIDIA-Omniverse/IsaacGymEnvs.git
   cd IsaacGymEnvs
   pip install -e .

5. Install rsl_rl

   git clone https://github.com/leggedrobotics/rsl_rl.git
   cd rsl_rl && git checkout v1.0.2
   pip install -e .

6. Clone this repository

   git clone https://github.com/wrlab/MotionPrediction.git

PoseLib

1. Create python virtual environment with python 3.7
2. Install Isaac Gym

   cd issacgym/python
   pip install -e .

3. Install Issac Gym Envs

   git clone https://github.com/NVIDIA-Omniverse/IsaacGymEnvs.git
   cd IsaacGymEnvs
   pip install -e .

4. Install bpy

   wget https://files.pythonhosted.org/packages/f4/fd/749f83e5006a3ecd2b134b20b42b7d5140e90d0ff3b45111abcb75027e80/bpy-2.91a0-cp37-cp37m-manylinux2014_x86_64.whl
   pip install bpy-2.91a0-cp37-cp37m-manylinux2014_x86_64.whl && bpy_post_install

5. Install FBX SDK
   • FBX SDK 2020.2.1 Python

   tar -zxvf fbx202021_fbxpythonsdk_linux.tar.gz
   chmod ugo+x fbx202021_fbxpythonsdk_linux ./fbx202021_fbxpythonsdk_linux ~/fbxsdk
   cp fbxsdk/lib/Python37_x64/* /anaconda3/envs/__name__/lib/python3.7/site-packages

Dataset

• CMU Mocap dataset