/nba-rookies

Facial recognition of NBA players

Primary LanguagePython

NBA's 2022-2023 season rookies facial recognition

This repository employs facial biometry models from InsightFace to perform facial recognition on NBA's 2023 best performing rookies, Victor Wembanyama and Scoot Henderson. Target subjects are compared among themselves as well as NBA star Ja Morant.

Setup

To respectively download models and create a Python virtual environment, run the scripts download_models.sh and install_requirements.sh.

Running demos

Be sure to include the repository's root in your Python interpreter's path list. For example, to run the face verification demo from the repository's root, execute:

PYTHONPATH=. .venv/bin/python demos/verification.py

Validating code quality

One can further statically analyze code for linting (flake8) and type checking (mypy) by installing libraries at requirements/dev.txt and running the script static_check.sh.