Human Tracking

Human tracking implementation in python using YOLOv3. Example video here: miss_dior_hd_output

Install

Clone

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

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:

make install

Sometimes the environment is not properly activated. Try to run:

conda activate aive

Models

Download YOLOv3 and tiny-YOLOv3 and place them in the models folder.

Run

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

make run

Otherwise, you can use the following command:

python src/main.py -i <VIDEO_PATH> -m <MODEL> -s <SPEED>

For the model choice, it can be:

• yolo
• tiny-yolo

For the speed, you can choose:

• normal
• fast
• faster
• flash

Development

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

brew install black
make format