first commit

.gitignore

@@ -0,0 +1,53 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# Caches
+.cache/
+*.cache
+
+# Virtual environment
+env/
+venv/
+ENV/
+.venv/
+
+# Local environment configurations
+.env
+*.env
+
+# Database files
+*.db
+*.sqlite3
+*.db-journal
+
+# Images, temporary files, and uploads
+/tmp/
+*.png
+*.jpg
+*.jpeg
+*.gif
+*.bmp
+
+# Logs
+*.log
+logs/
+
+# Jupyter Notebook checkpoints
+.ipynb_checkpoints/
+
+# Compiled C extensions
+*.so
+
+# macOS specific files
+.DS_Store
+
+# Windows specific files
+Thumbs.db
+ehthumbs.db
+Desktop.ini
+
+# System files
+.vscode/
+.idea/

app.py

@@ -0,0 +1,123 @@
+from fastapi import FastAPI, File, UploadFile, HTTPException, Form
+from deepface import DeepFace
+from manticoresearch import ApiClient, Configuration
+from manticoresearch.api import IndexApi, SearchApi
+from manticoresearch.model import SearchRequest
+from manticoresearch.model import InsertDocumentRequest
+import os
+import shutil
+from numpy import dot
+from numpy.linalg import norm
+
+# Configurar el cliente API
+config = Configuration()
+config.host = "http://localhost:9308"
+client = ApiClient(configuration=config)
+
+# Inicializar las APIs
+index_api = IndexApi(client)
+search_api = SearchApi(client)
+
+app = FastAPI()
+
+# Función para calcular similitud de coseno
+def cosine_similarity(vec1, vec2):
+    return dot(vec1, vec2) / (norm(vec1) * norm(vec2))
+
+# Ruta para registrar un nuevo usuario
+@app.post("/register_user")
+async def register_user(user_id: str = Form(...), image: UploadFile = File(...)):
+    try:
+        # Guardar la imagen temporalmente
+        img_path = f"/tmp/{image.filename}"
+        with open(img_path, "wb") as buffer:
+            shutil.copyfileobj(image.file, buffer)
+
+        # Obtener los embeddings faciales
+        embedding_result = DeepFace.represent(img_path, model_name="ArcFace")
+
+        # Extraer los embeddings
+        embeddings = embedding_result[0]['embedding']  # Extrae el vector de embeddings
+
+        # Verifica si embeddings tiene 512 valores
+        if len(embeddings) != 512:
+            raise HTTPException(status_code=400, detail=f"Los embeddings generados no tienen 512 valores: {len(embeddings)}")
+
+        # Almacenar los embeddings en Manticore Search
+        doc = {
+            "user_id": user_id,
+            "embeddings": embeddings
+        }
+
+        # Crea la solicitud de inserción de documento
+        insert_request = InsertDocumentRequest(index="users", doc=doc)
+
+        # Inserta el documento en Manticore Search
+        index_api.insert(insert_document_request=insert_request)
+
+        # Eliminar el archivo temporal
+        os.remove(img_path)
+
+        return {"status": "user registered successfully"}
+
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Error registering user: {str(e)}")
+
+# Ruta para verificar una imagen
+@app.post("/verify_user")
+async def verify_user(image: UploadFile = File(...)):
+    try:
+        # Guardar la imagen temporalmente
+        img_path = f"/tmp/{image.filename}"
+        with open(img_path, "wb") as buffer:
+            shutil.copyfileobj(image.file, buffer)
+
+        # Obtener los embeddings faciales de la imagen
+        query_embeddings = DeepFace.represent(img_path, model_name="ArcFace")[0]['embedding']
+
+        # Consulta en Manticore Search para obtener los usuarios registrados
+        query = {
+            "size": 100,  # Ajusta el tamaño si tienes muchos usuarios
+            "query": {
+                "match_all": {}
+            }
+        }
+
+        # Crear la solicitud de búsqueda
+        search_request = SearchRequest(index="users", query=query)
+
+        # Realizar la búsqueda
+        result = search_api.search(search_request)
+
+        # Recuperar todos los documentos (usuarios y sus embeddings)
+        hits = result.hits.hits
+        if not hits:
+            raise HTTPException(status_code=404, detail="No registered users found.")
+
+        # Calcular similitud de coseno con cada usuario registrado
+        max_similarity = -1
+        matched_user = None
+        for hit in hits:
+            stored_embeddings = hit['_source']['embeddings']  # Embeddings del usuario en la base de datos
+            similarity = cosine_similarity(query_embeddings, stored_embeddings)
+
+            # Imprimir la similitud de cada usuario para depuración
+            print(f"Similitud con {hit['_source']['user_id']}: {similarity}")
+
+            # Encontrar la mayor similitud
+            if similarity > max_similarity:
+                max_similarity = similarity
+                matched_user = hit['_source']['user_id']
+
+        # Establecer un umbral para decidir si hay coincidencia
+        similarity_threshold = 0.5  # Ajusta este valor según tus necesidades
+        if max_similarity < similarity_threshold:
+            raise HTTPException(status_code=404, detail="No matching user found")
+
+        # Eliminar el archivo temporal
+        os.remove(img_path)
+
+        return {"status": "success", "matched_user": matched_user, "similarity": max_similarity}
+
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"Error verifying user: {str(e)}")

migrations/pacedeep_fastapi.sql

@@ -0,0 +1,2 @@
+DROP TABLE IF EXISTS users;
+CREATE TABLE users (user_id text, embeddings float_vector KNN_TYPE='hnsw' KNN_DIMS='512' HNSW_SIMILARITY='cosine');

requirements.txt

@@ -0,0 +1,74 @@
+absl-py==2.1.0
+annotated-types==0.7.0
+anyio==4.6.2.post1
+astunparse==1.6.3
+beautifulsoup4==4.12.3
+blinker==1.8.2
+certifi==2024.8.30
+charset-normalizer==3.4.0
+click==8.1.7
+deepface==0.0.93
+exceptiongroup==1.2.2
+fastapi==0.115.2
+filelock==3.16.1
+fire==0.7.0
+Flask==3.0.3
+Flask-Cors==5.0.0
+flatbuffers==24.3.25
+gast==0.6.0
+gdown==5.2.0
+google-pasta==0.2.0
+grpcio==1.67.0
+gunicorn==23.0.0
+h11==0.14.0
+h5py==3.12.1
+idna==3.10
+itsdangerous==2.2.0
+Jinja2==3.1.4
+joblib==1.4.2
+keras==3.6.0
+libclang==18.1.1
+lz4==4.3.3
+manticoresearch==4.0.0
+Markdown==3.7
+markdown-it-py==3.0.0
+MarkupSafe==3.0.2
+mdurl==0.1.2
+ml-dtypes==0.4.1
+mtcnn==1.0.0
+namex==0.0.8
+numpy==1.26.4
+opencv-python==4.10.0.84
+opt_einsum==3.4.0
+optree==0.13.0
+packaging==24.1
+pandas==2.2.3
+pillow==11.0.0
+protobuf==4.25.5
+pydantic==2.9.2
+pydantic_core==2.23.4
+Pygments==2.18.0
+PySocks==1.7.1
+python-dateutil==2.9.0.post0
+python-multipart==0.0.12
+pytz==2024.2
+requests==2.32.3
+retina-face==0.0.17
+rich==13.9.2
+six==1.16.0
+sniffio==1.3.1
+soupsieve==2.6
+starlette==0.40.0
+tensorboard==2.17.1
+tensorboard-data-server==0.7.2
+tensorflow==2.17.0
+tensorflow-io-gcs-filesystem==0.37.1
+termcolor==2.5.0
+tf_keras==2.17.0
+tqdm==4.66.5
+typing_extensions==4.12.2
+tzdata==2024.2
+urllib3==2.2.3
+uvicorn==0.32.0
+Werkzeug==3.0.4
+wrapt==1.16.0