deepface-basic-usage

npx skills add majiayu000/claude-skill-registry --skill deepface-basic-usage --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

# Skill: DeepFace Basic Usage This skill provides a quick reference and guide for using the `deepface` library for face recognition tasks. ## Overview DeepFace is a hybrid face recognition framework for Python that wraps state-of-the-art models (VGG-Face, FaceNet, ArcFace, etc.) and detectors (RetinaFace, MtCNN, YOLO, etc.). ## Installation ```bash pip install deepface # Or with pixi pixi add deepface ``` ## Core Functions Import: ```python from deepface import DeepFace ``` ### 1. Face Verification (1:1) Determines if two images belong to the same person. ```python result = DeepFace.verify( img1_path = "img1.jpg", img2_path = "img2.jpg", model_name = "VGG-Face", # Optional: see Model Selection below detector_backend = "opencv", # Optional: see Detector Selection below distance_metric = "cosine", # Optional: cosine, euclidean, euclidean_l2 align = True, # Optional: Align faces before verification ) # result is a dict: # { # "verified": True, # "distance": 0.256, # "threshold": 0.40, # "model": "VGG-Face", # "detector_backend": "opencv", # "similarity_metric": "cosine", # "facial_areas": {"img1": {...}, "img2": {...}}, # "time": 1.23 # } ``` ### 2. Face Recognition / Search (1:N) Fin

pip install deepface
Or with pixi
pixi add deepface