🔀 [Merge] branch 'main' into HF

yolo/tools/data_augmentation.py

@@ -67,7 +67,7 @@ class PadAndResize:
         scale = min(self.target_width / img_width, self.target_height / img_height)
         new_width, new_height = int(img_width * scale), int(img_height * scale)
 
-        resized_image = image.resize((new_width, new_height), Image.LANCZOS)
+        resized_image = image.resize((new_width, new_height), Image.Resampling.LANCZOS)
 
         pad_left = (self.target_width - new_width) // 2
         pad_top = (self.target_height - new_height) // 2

yolo/tools/data_loader.py

@@ -126,10 +126,12 @@ class YoloDataset(Dataset):
 
     def load_valid_labels(self, label_path: str, seg_data_one_img: list) -> Union[Tensor, None]:
         """
-        Loads and validates bounding box data is [0, 1] from a label file.
+        Loads valid COCO style segmentation data (values between [0, 1]) and converts it to bounding box coordinates
+        by finding the minimum and maximum x and y values.
 
         Parameters:
-            label_path (str): The filepath to the label file containing bounding box data.
+            label_path (str): The filepath to the label file containing annotation data.
+            seg_data_one_img (list): The actual list of annotations (in segmentation format)
 
         Returns:
             Tensor or None: A tensor of all valid bounding boxes if any are found; otherwise, None.

yolo/tools/solver.py

@@ -48,17 +48,9 @@ class ValidateModel(BaseModel):
         batch_size, images, targets, rev_tensor, img_paths = batch
         H, W = images.shape[2:]
         predicts = self.post_process(self.ema(images), image_size=[W, H])
-        batch_metrics = self.metric(
+        self.metric.update(
             [to_metrics_format(predict) for predict in predicts], [to_metrics_format(target) for target in targets]
         )
-
-        self.log_dict(
-            {
-                "map": batch_metrics["map"],
-                "map_50": batch_metrics["map_50"],
-            },
-            batch_size=batch_size,
-        )
         return predicts
 
     def on_validation_epoch_end(self):

yolo/utils/bounding_box_utils.py

@@ -484,6 +484,7 @@ def calculate_map(predictions, ground_truths) -> Dict[str, Tensor]:
 
 
 def to_metrics_format(prediction: Tensor) -> Dict[str, Union[float, Tensor]]:
+    prediction = prediction[prediction[:, 0] != -1]
     bbox = {"boxes": prediction[:, 1:5], "labels": prediction[:, 0].int()}
     if prediction.size(1) == 6:
         bbox["scores"] = prediction[:, 5]

yolo/utils/dataset_utils.py

@@ -104,7 +104,9 @@ def scale_segmentation(
         if "segmentation" in anno:
             seg_list = [item for sublist in anno["segmentation"] for item in sublist]
         elif "bbox" in anno:
-            seg_list = anno["bbox"]
+            x, y, width, height = anno["bbox"]
+            seg_list = [x, y, x + width, y, x + width, y + height, x, y + height]
+
         scaled_seg_data = (
             np.array(seg_list).reshape(-1, 2) / [w, h]
         ).tolist()  # make the list group in x, y pairs and scaled with image width, height