Files
SnapOtter/packages/ai/python/detect_faces.py
T

318 lines
11 KiB
Python
Raw Normal View History

"""Face detection and blurring using MediaPipe."""
import sys
import json
import os
def emit_progress(percent, stage):
"""Emit structured progress to stderr for bridge.ts to capture."""
print(json.dumps({"progress": percent, "stage": stage}), file=sys.stderr, flush=True)
# ── Model path for new mp.tasks API ─────────────────────────────────
_MODELS_BASE = os.environ.get("MODELS_PATH", "/opt/models")
_FACE_DETECT_MODEL_URL = "https://storage.googleapis.com/mediapipe-models/face_detector/blaze_face_short_range/float16/latest/blaze_face_short_range.tflite"
_DOCKER_MODEL_PATH = os.path.join(_MODELS_BASE, "mediapipe", "blaze_face_short_range.tflite")
_LOCAL_MODEL_DIR = os.path.join(os.path.dirname(__file__), "..", "..", "..", ".models")
_LOCAL_MODEL_PATH = os.path.join(_LOCAL_MODEL_DIR, "blaze_face_short_range.tflite")
def _ensure_face_detect_model():
"""Resolve face detector model. Docker path first, then local dev."""
if os.path.exists(_DOCKER_MODEL_PATH):
return _DOCKER_MODEL_PATH
if os.path.exists(_LOCAL_MODEL_PATH):
return _LOCAL_MODEL_PATH
os.makedirs(_LOCAL_MODEL_DIR, exist_ok=True)
import urllib.request
emit_progress(15, "Downloading face detection model")
urllib.request.urlretrieve(_FACE_DETECT_MODEL_URL, _LOCAL_MODEL_PATH)
return _LOCAL_MODEL_PATH
def _iou(a, b):
"""Compute intersection-over-union between two face boxes."""
ax2, ay2 = a["x"] + a["w"], a["y"] + a["h"]
bx2, by2 = b["x"] + b["w"], b["y"] + b["h"]
inter_w = max(0, min(ax2, bx2) - max(a["x"], b["x"]))
inter_h = max(0, min(ay2, by2) - max(a["y"], b["y"]))
inter = inter_w * inter_h
union = a["w"] * a["h"] + b["w"] * b["h"] - inter
return inter / union if union > 0 else 0.0
def _nms_faces(faces, iou_threshold=0.4):
"""Remove duplicate detections using greedy non-maximum suppression."""
if len(faces) <= 1:
return faces
kept = []
used = [False] * len(faces)
for i in range(len(faces)):
if used[i]:
continue
kept.append(faces[i])
used[i] = True
for j in range(i + 1, len(faces)):
if not used[j] and _iou(faces[i], faces[j]) >= iou_threshold:
used[j] = True
return kept
def _detect_with_solutions(img_array, min_confidence):
"""Detect faces using legacy mp.solutions API (mediapipe < 0.10.30).
Runs both short-range (model 0) and full-range (model 1) detectors and
merges the results. Previously the loop broke on the first model that
found any face, so group photos where model 0 caught only 1-2 large
faces would never have the remaining faces scanned by model 1.
"""
import mediapipe as mp
mp_face = mp.solutions.face_detection
ih, iw = img_array.shape[:2]
all_faces = []
for model_sel in [0, 1]:
detector = mp_face.FaceDetection(
model_selection=model_sel,
min_detection_confidence=min_confidence,
)
results = detector.process(img_array)
detector.close()
for detection in (results.detections or []):
bbox = detection.location_data.relative_bounding_box
all_faces.append({
"x": int(bbox.xmin * iw),
"y": int(bbox.ymin * ih),
"w": int(bbox.width * iw),
"h": int(bbox.height * ih),
})
return _nms_faces(all_faces)
def _detect_with_tasks(img_array, min_confidence):
"""Detect faces using new mp.tasks API (mediapipe >= 0.10.30)."""
import mediapipe as mp
model_path = _ensure_face_detect_model()
options = mp.tasks.vision.FaceDetectorOptions(
base_options=mp.tasks.BaseOptions(model_asset_path=model_path),
running_mode=mp.tasks.vision.RunningMode.IMAGE,
min_detection_confidence=min_confidence,
)
detector = mp.tasks.vision.FaceDetector.create_from_options(options)
mp_image = mp.Image(image_format=mp.ImageFormat.SRGB, data=img_array)
result = detector.detect(mp_image)
detector.close()
faces = []
for detection in result.detections:
bbox = detection.bounding_box
faces.append({
"x": bbox.origin_x,
"y": bbox.origin_y,
"w": bbox.width,
"h": bbox.height,
})
return faces
_MAX_DETECT_DIM = 1920
def _downscale_for_detection(img_array):
"""Downscale image if needed so MediaPipe can detect faces reliably.
Returns (scaled_array, scale_factor). Coordinates from detection on
the scaled image must be multiplied by scale_factor to map back to
the original resolution.
"""
h, w = img_array.shape[:2]
longest = max(h, w)
if longest <= _MAX_DETECT_DIM:
return img_array, 1.0
import cv2
scale = _MAX_DETECT_DIM / longest
new_w = int(w * scale)
new_h = int(h * scale)
resized = cv2.resize(img_array, (new_w, new_h), interpolation=cv2.INTER_AREA)
return resized, 1.0 / scale
def _detect_single_orientation(img_array, min_confidence):
"""Run face detection on a single image orientation."""
try:
return _detect_with_solutions(img_array, min_confidence)
except AttributeError:
return _detect_with_tasks(img_array, min_confidence)
def _remap_faces_from_rotation(faces, rotation, orig_h, orig_w):
"""Map face coordinates from a rotated image back to the original.
rotation: 90, 180, or 270 (clockwise degrees applied to the original).
orig_h, orig_w: dimensions of the original (un-rotated) image.
When the original is (orig_h, orig_w):
90 CW -> rotated is (orig_w, orig_h). (rx, ry) -> (ry, orig_h - rx - rw_box)
180 -> rotated is (orig_h, orig_w). (rx, ry) -> (orig_w - rx - rw_box, orig_h - ry - rh_box)
270 CW -> rotated is (orig_w, orig_h). (rx, ry) -> (orig_w - ry - rh_box, rx)
"""
remapped = []
for f in faces:
x, y, w, h = f["x"], f["y"], f["w"], f["h"]
if rotation == 90:
remapped.append({"x": y, "y": orig_h - x - w, "w": h, "h": w})
elif rotation == 180:
remapped.append({"x": orig_w - x - w, "y": orig_h - y - h, "w": w, "h": h})
elif rotation == 270:
remapped.append({"x": orig_w - y - h, "y": x, "w": h, "h": w})
else:
remapped.append(f)
return remapped
def _detect_faces(img_array, min_confidence):
"""Detect faces, trying multiple orientations if needed.
MediaPipe BlazeFace can miss faces in portrait-oriented or rotated
images. We first try the image as-is; if no faces are found we
retry at 90, 180, and 270-degree rotations and map the coordinates
back. Large images are downscaled before detection.
"""
import cv2
orig_h, orig_w = img_array.shape[:2]
scaled, inv_scale = _downscale_for_detection(img_array)
faces = _detect_single_orientation(scaled, min_confidence)
# If nothing found, try rotated copies and merge all detections.
# Different rotations can catch different faces, so we union them
# and de-duplicate with NMS.
if not faces:
rotations = [
(cv2.ROTATE_90_CLOCKWISE, 90),
(cv2.ROTATE_180, 180),
(cv2.ROTATE_90_COUNTERCLOCKWISE, 270),
]
all_rotated_faces = []
sh, sw = scaled.shape[:2]
for cv2_flag, degrees in rotations:
rotated = cv2.rotate(scaled, cv2_flag)
found = _detect_single_orientation(rotated, min_confidence)
if found:
remapped = _remap_faces_from_rotation(found, degrees, sh, sw)
all_rotated_faces.extend(remapped)
faces = _nms_faces(all_rotated_faces)
if inv_scale != 1.0:
for f in faces:
f["x"] = int(f["x"] * inv_scale)
f["y"] = int(f["y"] * inv_scale)
f["w"] = int(f["w"] * inv_scale)
f["h"] = int(f["h"] * inv_scale)
return faces
def main():
input_path = sys.argv[1]
output_path = sys.argv[2]
settings = json.loads(sys.argv[3]) if len(sys.argv) > 3 else {}
blur_radius = settings.get("blurRadius", 30)
sensitivity = settings.get("sensitivity", 0.5)
detect_only = settings.get("detectOnly", False)
try:
emit_progress(10, "Preparing")
from PIL import Image, ImageFilter
img = Image.open(input_path).convert("RGB")
try:
import numpy as np
emit_progress(20, "Ready")
# Map sensitivity (0.1-0.9) to MediaPipe confidence threshold.
# Higher sensitivity = lower confidence threshold = more detections.
min_confidence = max(0.1, 1.0 - sensitivity)
img_array = np.array(img)
# Try legacy mp.solutions API first, fall back to mp.tasks
emit_progress(25, "Scanning for faces")
faces = _detect_faces(img_array, min_confidence)
num_faces = len(faces)
emit_progress(50, f"Found {num_faces} face{'s' if num_faces != 1 else ''}")
if num_faces > 0 and not detect_only:
for i, face in enumerate(faces):
x, y, w, h = face["x"], face["y"], face["w"], face["h"]
# Add padding around the face
pad = int(max(w, h) * 0.1)
x1 = max(0, x - pad)
y1 = max(0, y - pad)
x2 = min(img.width, x + w + pad)
y2 = min(img.height, y + h + pad)
face_region = img.crop((x1, y1, x2, y2))
blurred = face_region.filter(
ImageFilter.GaussianBlur(blur_radius)
)
img.paste(blurred, (x1, y1))
emit_progress(
50 + int((i + 1) / num_faces * 40),
f"Blurring face {i + 1} of {num_faces}",
)
if not detect_only:
emit_progress(95, "Saving result")
img.save(output_path)
print(
json.dumps(
{
"success": True,
"facesDetected": len(faces),
"faces": faces,
}
)
)
except ImportError:
print(
json.dumps(
{
"success": False,
"error": "Face detection requires MediaPipe. Install with: pip install mediapipe",
}
)
)
sys.exit(1)
except ImportError:
print(
json.dumps(
{
"success": False,
"error": "Pillow is not installed. Install with: pip install Pillow",
}
)
)
sys.exit(1)
except Exception as e:
print(json.dumps({"success": False, "error": str(e)}))
sys.exit(1)
if __name__ == "__main__":
main()