Update app.py

app.py

@@ -39,12 +39,26 @@ if api_key:
 
     def segment_image(image_path):
         try:
-            results = CLIENT.infer(image_path, model_id="closet/1")
+            # 이미지 파일 읽기
+            with open(image_path, "rb") as image_file:
+                image_data = image_file.read()
             
+            # 이미지를 base64로 인코딩
+            encoded_image = base64.b64encode(image_data).decode('utf-8')
+            
+            # 원본 이미지 로드 (여기로 이동)
             image = cv2.imread(image_path)
             image = cv2.resize(image, (800, 600))
             mask = np.zeros(image.shape, dtype=np.uint8)
             
+            try:
+                # Roboflow API 호출
+                results = CLIENT.infer(encoded_image, model_id="closet/1")
+            except HTTPCallErrorError as api_error:
+                st.error(f"API Error: {api_error.description}")
+                st.error(f"API Message: {api_error.api_message}")
+                return Image.fromarray(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))
+            
             if 'predictions' in results:
                 for prediction in results['predictions']:
                     points = prediction['points']
@@ -55,7 +69,7 @@ if api_key:
                     pts = pts.astype(np.int32)
                     pts = pts.reshape((-1, 1, 2))
                     cv2.fillPoly(mask, [pts], color=(255, 255, 255))  # White mask
-            
+                
                 segmented_image = cv2.bitwise_and(image, mask)
             else:
                 st.warning("No predictions found in the image. Returning original image.")
@@ -64,8 +78,8 @@ if api_key:
             return Image.fromarray(cv2.cvtColor(segmented_image, cv2.COLOR_BGR2RGB))
         except Exception as e:
             st.error(f"Error in segmentation: {str(e)}")
-            return Image.open(image_path)  # Return original image if segmentation fails
-
+            # 원본 이미지를 다시 읽어 반환
+            return Image.open(image_path)
     def get_image_embedding(image):
         image_tensor = preprocess_val(image).unsqueeze(0).to(device)
         with torch.no_grad():