Spaces:
Sleeping
Sleeping
File size: 3,632 Bytes
9bc686b | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 | """
Re-enrollment script: Re-generates face embeddings from existing enrolled images
using the current face engine. Run this after upgrading the face engine.
Usage:
..\.venv\Scripts\python.exe re_enroll.py
"""
import sys
import os
# Add the backend to path
sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
# Set env vars before importing app modules
from dotenv import load_dotenv
load_dotenv()
from app.core.database import SessionLocal
from app.models import models
from app.services.singletons import face_engine
import cv2
import numpy as np
import logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger("ReEnroll")
def re_enroll():
db = SessionLocal()
try:
# Get all employees with images
employees = db.query(models.Employee).all()
logger.info(f"Found {len(employees)} employees to re-enroll.")
for employee in employees:
images = db.query(models.EmployeeImage).filter(
models.EmployeeImage.employee_id == employee.id
).all()
if not images:
logger.warning(f"No images found for employee {employee.name} ({employee.employee_id})")
continue
logger.info(f"\nRe-enrolling {employee.name} ({employee.employee_id}) - {len(images)} pose(s)...")
# Delete existing embeddings for this employee
db.query(models.FaceEmbedding).filter(
models.FaceEmbedding.employee_id == employee.id
).delete()
db.commit()
re_enrolled_count = 0
for img_record in images:
# Read image from disk
if not os.path.exists(img_record.file_path):
logger.warning(f" Image file not found: {img_record.file_path}")
continue
img = cv2.imread(img_record.file_path)
if img is None:
logger.warning(f" Failed to read image: {img_record.file_path}")
continue
# Detect face
faces = face_engine.detect_faces(img)
if not faces:
logger.warning(f" No face detected in {img_record.file_path}")
continue
face = faces[0]
# Align and extract embedding
aligned = face_engine.align_face(img, face["landmarks"])
embedding = face_engine.extract_embedding(aligned)
embedding_list = embedding.tolist()
# Save new embedding
new_emb = models.FaceEmbedding(
employee_id=employee.id,
image_id=img_record.id,
embedding=embedding_list
)
db.add(new_emb)
re_enrolled_count += 1
logger.info(f" ✓ Re-enrolled pose: {img_record.pose_type}")
db.commit()
logger.info(f" Total re-enrolled: {re_enrolled_count}/{len(images)} poses for {employee.name}")
# Final count
total_embs = db.query(models.FaceEmbedding).count()
logger.info(f"\n✅ Re-enrollment complete! Total embeddings in DB: {total_embs}")
except Exception as e:
logger.error(f"Re-enrollment failed: {e}")
import traceback
traceback.print_exc()
db.rollback()
finally:
db.close()
if __name__ == "__main__":
re_enroll()
|