CMU MoCap Data Loader - Setup Guide

This project loads and visualizes motion capture data from the CMU enge817_dataset.

Project Structure

• load_mocap.ipynb - Main Jupyter notebook for loading and analyzing MoCap data
• environment.yml - Conda environment configuration
• mocap_preview.png - Generated visualization of joint angles

Dataset

• Location: C:\Users\harma\Downloads\enge817_dataset
• Subjects: 12, 17, 22, 35, 41
• File Formats:
  • .asf - Skeleton structure (joint names, bone lengths, hierarchy)
  • .amc - Motion sequences (joint angles at 120 Hz)

Setup Instructions

Option A: Create Environment from environment.yml (Recommended)

This approach ensures reproducibility and is the easiest method.

# Create environment from environment.yml
conda env create -f environment.yml

# Activate the environment
conda activate enge817

# Register the kernel with Jupyter
python -m ipykernel install --user --name enge817 --display-name "Python (enge817)"

Verify Installation

After setup, verify everything is working correctly:

# Check that packages are installed
conda list

# List all available conda environments
conda env list

# List all available Jupyter kernels
jupyter kernelspec list

You should see enge817 listed in both the conda environments and Jupyter kernels.

To Update Later

If you need to add or update packages, follow these steps:

# Activate the environment
conda activate enge817

# Install a new package (example)
conda install -c conda-forge <package-name>

# Export updated environment to environment.yml
conda env export --from-history > environment.yml

# Commit the updated environment.yml to version control

Usage

1. Activate the environment:

   conda activate enge817

2. Start Jupyter:

   jupyter lab
   # or
   jupyter notebook

3. Open load_mocap.ipynb and select the enge817 kernel from the kernel selector

4. Run cells sequentially or jump to specific analyses

Features

• Parse ASF skeleton structure and extract joint hierarchy
• Load AMC motion sequences (joint angle time series)
• Generate motion data array: shape [T, J] where T=frames, J=DOF
• Visualize joint angles over time for lower limb joints
• Compute motion statistics and range of motion analysis
• Support for multiple subjects and motion sequences

Dependencies

• Python 3.10
• numpy - Numerical computing
• matplotlib - Plotting and visualization
• jupyter/jupyterlab - Interactive notebooks
• ipykernel - Kernel for Jupyter integration

Troubleshooting

Kernel not showing in Jupyter:

conda activate enge817
python -m ipykernel install --user --name enge817 --display-name "Python (enge817)"

Need to remove the environment:

conda remove -n enge817 --all

Update conda:

conda update conda