fix: raise detection confidence threshold to 0.7, move model cache to project dir

.gitignore

@@ -0,0 +1,13 @@
+# Python-generated files
+__pycache__/
+*.py[oc]
+build/
+dist/
+wheels/
+*.egg-info
+
+# Virtual environments
+.venv
+
+# Downloaded ML models
+models/

src/faceblur/cli.py

@@ -1,9 +1,16 @@
 """CLI module for faceblur-poc."""
 
 import argparse
+import os
 import sys
 from pathlib import Path
 
+# Set model cache to project-local models/ directory before importing uniface
+os.environ.setdefault(
+    "UNIFACE_CACHE_DIR",
+    str(Path(__file__).resolve().parent.parent.parent / "models"),
+)
+
 from .video import extract_frames
 from .detect import FaceDetector
 from .cluster import cluster_faces
@@ -35,6 +42,12 @@ def main():
     detect_parser.add_argument(
         "--min-samples", type=int, default=2, help="DBSCAN min_samples"
     )
+    detect_parser.add_argument(
+        "--confidence",
+        type=float,
+        default=0.7,
+        help="Minimum face detection confidence (0-1)",
+    )
 
     args = parser.parse_args()
 
@@ -66,7 +79,7 @@ def run_detect(args):
         sys.exit(1)
 
     print("\n[2/5] Initializing face detector...")
-    detector = FaceDetector()
+    detector = FaceDetector(confidence_threshold=args.confidence)
 
     print("\n[3/5] Detecting faces...")
     all_faces = []

src/faceblur/detect.py

@@ -26,8 +26,8 @@ class FaceData:
 class FaceDetector:
     """Face detector using RetinaFace + ArcFace via UniFace."""
 
-    def __init__(self):
+    def __init__(self, confidence_threshold: float = 0.7):
-        self.detector = RetinaFace()
+        self.detector = RetinaFace(confidence_threshold=confidence_threshold)
         self.recognizer = ArcFace()
 
     def detect_faces(self, frame_path: Path, frame_index: int) -> List[FaceData]: