Update app.py

app.py

@@ -15,13 +15,14 @@ depth_model.eval()
 def estimate_depth(image):
     """Estimate depth map from image."""
     image = image.convert("RGB")
+    image = image.resize((384, 384))  # Resize for model input
     inputs = feature_extractor(images=image, return_tensors="pt")
     with torch.no_grad():
         outputs = depth_model(**inputs)
         depth = outputs.predicted_depth.squeeze().cpu().numpy()
+    depth = cv2.resize(depth, (image.width, image.height))  # Resize back to original
     depth = (depth - depth.min()) / (depth.max() - depth.min()) * 255
-    depth = depth.astype(np.uint8)
-    return cv2.GaussianBlur(depth, (5, 5), 0)  # Smooth depth map to reduce noise
+    return depth.astype(np.uint8)
 
 def warp_design(cloth_img, design_img):
     """Warp the design onto the clothing while preserving folds."""
@@ -30,19 +31,20 @@ def warp_design(cloth_img, design_img):
     cloth_np = np.array(cloth_img)
     design_np = np.array(design_img)
     
-    # Ensure both images have the same dimensions
+    # Resize design to match clothing dimensions
     design_np = cv2.resize(design_np, (cloth_np.shape[1], cloth_np.shape[0]))
     
     # Estimate depth for fold detection
     depth_map = estimate_depth(cloth_img)
+    depth_map = cv2.resize(depth_map, (cloth_np.shape[1], cloth_np.shape[0]))  # Ensure matching dimensions
     
     # Generate displacement map based on depth
-    displacement_x = cv2.Sobel(depth_map, cv2.CV_32F, 1, 0, ksize=3)
-    displacement_y = cv2.Sobel(depth_map, cv2.CV_32F, 0, 1, ksize=3)
+    displacement_x = cv2.Sobel(depth_map, cv2.CV_32F, 1, 0, ksize=5)
+    displacement_y = cv2.Sobel(depth_map, cv2.CV_32F, 0, 1, ksize=5)
     
     # Normalize displacement values
-    displacement_x = cv2.normalize(displacement_x, None, -3, 3, cv2.NORM_MINMAX)
-    displacement_y = cv2.normalize(displacement_y, None, -3, 3, cv2.NORM_MINMAX)
+    displacement_x = cv2.normalize(displacement_x, None, -5, 5, cv2.NORM_MINMAX)
+    displacement_y = cv2.normalize(displacement_y, None, -5, 5, cv2.NORM_MINMAX)
     
     # Warp design using displacement map
     h, w, _ = cloth_np.shape
@@ -51,11 +53,8 @@ def warp_design(cloth_img, design_img):
     map_y = np.clip(np.float32(map_y + displacement_y), 0, h - 1)
     warped_design = cv2.remap(design_np, map_x, map_y, interpolation=cv2.INTER_LINEAR, borderMode=cv2.BORDER_REFLECT)
     
-    # Convert warped design to match original image dtype
-    warped_design = warped_design.astype(np.uint8)
-    
     # Blend images
-    blended = cv2.addWeighted(cloth_np, 0.7, warped_design, 0.3, 0)
+    blended = cv2.addWeighted(cloth_np, 0.6, warped_design, 0.4, 0)
     return Image.fromarray(blended)
 
 def main(cloth, design):
@@ -70,4 +69,4 @@ iface = gr.Interface(
 )
 
 if __name__ == "__main__":
-    iface.launch()
+    iface.launch(share=True)