Update app.py

app.py

@@ -5,7 +5,6 @@ import torch
 from torchvision import transforms
 import cv2
 import numpy as np
-from facenet_pytorch import MTCNN
 
 # Function to load the ViT model
 def load_model(model_path):
@@ -14,26 +13,23 @@ def load_model(model_path):
     model.to(device)
     return model, device
 
-# Initialize MTCNN for face detection
-device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
-mtcnn = MTCNN(keep_all=True, device=device)
+# Load Haar Cascade for face detection
+face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
 
-# Function to preprocess the image using MTCNN for face detection
+# Function to preprocess the image using Haar Cascade for face detection
 def preprocess_image(image, device):
     # Convert PIL image to OpenCV format
     open_cv_image = np.array(image.convert("RGB"))
     # Convert RGB to BGR for OpenCV
     open_cv_image = cv2.cvtColor(open_cv_image, cv2.COLOR_RGB2BGR)
 
-    # Convert OpenCV image back to PIL Image for MTCNN
-    pil_image = Image.fromarray(cv2.cvtColor(open_cv_image, cv2.COLOR_BGR2RGB))
+    # Detect faces using Haar Cascade
+    faces = face_cascade.detectMultiScale(open_cv_image, scaleFactor=1.1, minNeighbors=5, minSize=(30, 30))
 
-    # Use MTCNN to detect faces
-    boxes, _ = mtcnn.detect(pil_image)
-    if boxes is not None:
+    if len(faces) > 0:
         # Crop the first detected face (for simplicity)
-        box = boxes[0].astype(int)
-        cropped_face = open_cv_image[box[1]:box[3], box[0]:box[2]]
+        (x, y, w, h) = faces[0]
+        cropped_face = open_cv_image[y:y+h, x:x+w]
         # Convert cropped face back to PIL for further processing
         processed_image = Image.fromarray(cv2.cvtColor(cropped_face, cv2.COLOR_BGR2RGB))
     else: