How bahrombekk/face_recognition_backup_full works
Project type and purpose
• This is a specialized face-recognition pipeline/CLI application in Python.
• It collects faces via camera, aligns/encodes them into 512-dim vectors using InsightFace, stores them in LMDB, and performs real-time recognition or single-image matching.
File structure
• Top level:
• faceid/ directory – contains all submodules (data, database, code, README).
• main.py – launcher stub (delegates to faceid/main.py).
• faceid/data/ – raw, aligned, and temp images.
• faceid/baza/ – LMDB database, backups, logs.
• faceid/cod/ – collector, config, encoder, recognition, tests, comparison script, main script.
README insights
• Core components: data collection, encoding, storage, recognition, comparison.
• Tech stack: Python3.8+, OpenCV, NumPy, InsightFace, TensorFlow, LMDB, scikit-learn.
• Supports CPU/GPU automatically.
• Configurable via cod/config/config.py.
• Logging under baza/logs/, tests under cod/tests/.