Update utils.py

utils.py

@@ -375,22 +375,22 @@ def predict(model, image, caption, box_threshold, text_threshold):
 
 
 def predict_yolo(model, image_path, box_threshold, imgsz, scale_img, iou_threshold=0.7):
-    """ Use huggingface model to replace the original model
-    """
-    # model = model['model']
-
+    """Use YOLO model for object detection with correct parameters"""
+    
+    # Should be:
     kwargs = {
-        'conf': box_threshold,
-        'iou': iou_threshold,
+        'conf_thres': box_threshold,  # Correct parameter name
+        'iou_thres': iou_threshold,
+        'verbose': False
     }
+    
     if scale_img:
         kwargs['imgsz'] = imgsz
-    result = model(image_path, **kwargs)
-    boxes = result[0].boxes.xyxy#.tolist() # in pixel space
-    conf = result[0].boxes.conf
-    phrases = [str(i) for i in range(len(boxes))]
-
-    return boxes, conf, phrases
+        
+    results = model.predict(image_path, **kwargs)
+    boxes = results[0].boxes.xyxy
+    conf = results[0].boxes.conf
+    return boxes, conf, [str(i) for i in range(len(boxes))]
 
 
 def get_som_labeled_img(img_path, model=None, BOX_TRESHOLD = 0.01, output_coord_in_ratio=False, ocr_bbox=None, text_scale=0.4, text_padding=5, draw_bbox_config=None, caption_model_processor=None, ocr_text=[], use_local_semantics=True, iou_threshold=0.9,prompt=None, scale_img=False, imgsz=None, batch_size=None):