Human Tracking

Human tracking implementation in python using QDTrack. Example video here: miss_dior_hd_output

Install

Clone

git clone [email protected]:Baiame/aive_test_mmtracking.git
cd aive_test_mmtracking

Python

Have Python >=3.9 installed with conda. Have make installed. For example using brew:

/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
)
brew install make

Install requirements: Follow the tutorial here: MMTRACKING installation The folder architecture should be:

.
├── Makefile
├── Readme.md
├── demo
├── input_videos
├── mmcv
├── mmtracking
└── output_videos

Models

Download QDTrack Faster R-CNN and place it in the ./demo/checkpoints folder.

Run

To run the test with the default video and default mode, run:

make run-hd

Otherwise, you can use the following command:

python ./demo/demo_vid.py \
    <MODEL_CONFIG>  \
    --input <INPUT_VIDEO> \
    --checkpoint <MODEL>  \
    --output <OUTPUT_VIDEO> \
    --show --device <DEVICE>

For the device, you can choose:

• cpu
• cuda (if you have a cuda compatible gpu)

Known errors:

ImportError: numpy.core.multiarray failed to import

Just update numpy:

pip install -U numpy 

Development

For code formatting, install black and run the formatting with make:

brew install black
make format