Update app.py

app.py

@@ -5,6 +5,13 @@
 
 
 
+# -*- coding: utf-8 -*-
+
+# Commented out IPython magic to ensure Python compatibility.
+
+
+
+
 import os
 import supervision as sv
 from PIL import Image, ImageFilter
@@ -845,7 +852,8 @@ def apply_mosaic(image, bbox, mosaic_size=10):
     face_region = cv2.resize(face_region, (w, h), interpolation=cv2.INTER_NEAREST)
     image[y:y+h, x:x+w] = face_region
     return image
-
+from fastapi import FastAPI, File, UploadFile, HTTPException
+from io import BytesIO  #
 @app.post("/mosaic_faces")
 async def mosaic_faces(reference_image: UploadFile = File(...), test_image: UploadFile = File(...)):
     try:
@@ -856,41 +864,30 @@ async def mosaic_faces(reference_image: UploadFile = File(...), test_image: Uplo
         # Extract face data from reference image
         reference_embeddings, _, _ = get_face_data(ref_image)
         if reference_embeddings is None:
-            return JSONResponse(status_code=400, content={"error": "No face detected in reference image."})
+            raise HTTPException(status_code=400, detail="No face detected in reference image.")
 
         # Extract face data from test image
         test_embeddings, test_bboxes, test_image_processed = get_face_data(test_image_data)
         if test_embeddings is None:
-            return JSONResponse(status_code=400, content={"error": "No face detected in test image."})
+            raise HTTPException(status_code=400, detail="No face detected in test image.")
 
         # Process each detected face in the test image
-        recognized = False
         for test_embedding, bbox in zip(test_embeddings, test_bboxes):
-            # Calculate cosine similarity between embeddings
             similarity = np.dot(reference_embeddings[0], test_embedding) / (
                 np.linalg.norm(reference_embeddings[0]) * np.linalg.norm(test_embedding))
 
-            # Check if the face is recognized based on similarity
-            if similarity > 0.4:
-                recognized = True
-                cv2.rectangle(test_image_processed, (int(bbox[0]), int(bbox[1])),
-                              (int(bbox[2]), int(bbox[3])), (0, 255, 0), 2)  # Green box for recognized face
-            else:
-                # Apply mosaic to other faces
+            # Apply mosaic to unrecognized faces only
+            if similarity <= 0.4:
                 test_image_processed = apply_mosaic(test_image_processed, bbox)
 
-        if not recognized:
-            return JSONResponse(status_code=200, content={"message": "No match found in the test image."})
+        # Save processed image to a temporary file
+        temp_file = "/tmp/processed_image.jpg"
+        cv2.imwrite(temp_file, test_image_processed)
 
-        # Encode the processed image to return as a response
-        _, buffer = cv2.imencode('.jpg', test_image_processed)
-        return JSONResponse(status_code=200, content={"message": "Faces processed successfully."})
+        return FileResponse(temp_file, media_type="image/jpeg", filename="processed_image.jpg")
 
     except Exception as e:
-        return JSONResponse(status_code=500, content={"error": str(e)})
-
-
-
+        raise HTTPException(status_code=500, detail=str(e))
 
 @app.get("/", response_class=HTMLResponse)
 async def read_root():
@@ -1072,7 +1069,5 @@ async def read_root():
 
     """
     return HTMLResponse(content=html_content)
-if __name__ == "__main__":
 
-    app.run(host="0.0.0.0", port=7860)