Image Path as keys for DetectionDataset

supervision/dataset/core.py

@@ -181,21 +181,21 @@ def as_pascal_voc(
             in the range [0, 1). Argument is used only for segmentation datasets.
         """
         if images_directory_path:
-            images_path = Path(images_directory_path)
-            images_path.mkdir(parents=True, exist_ok=True)
-
+            save_dataset_images(
+                images_directory_path=images_directory_path, images=self.images
+            )
         if annotations_directory_path:
-            annotations_path = Path(annotations_directory_path)
-            annotations_path.mkdir(parents=True, exist_ok=True)
+            Path(annotations_directory_path).mkdir(parents=True, exist_ok=True)
 
-        for image_name, image in self.images.items():
-            detections = self.annotations[image_name]
-
-            if images_directory_path:
-                cv2.imwrite(str(images_path / image_name), image)
+        for image_path, image in self.images.items():
+            detections = self.annotations[image_path]
 
             if annotations_directory_path:
-                annotation_name = Path(image_name).stem
+                annotation_name = Path(image_path).stem
+                annotations_path = os.path.join(
+                    annotations_directory_path, f"{annotation_name}.xml"
+                )
+                image_name = Path(image_path).name
                 pascal_voc_xml = detections_to_pascal_voc(
                     detections=detections,
                     classes=self.classes,
@@ -206,7 +206,7 @@ def as_pascal_voc(
                     approximation_percentage=approximation_percentage,
                 )
 
-                with open(annotations_path / f"{annotation_name}.xml", "w") as f:
+                with open(annotations_path, "w") as f:
                     f.write(pascal_voc_xml)
 
     @classmethod
@@ -615,9 +615,10 @@ def as_folder_structure(self, root_directory_path: str) -> None:
         for class_name in self.classes:
             os.makedirs(os.path.join(root_directory_path, class_name), exist_ok=True)
 
-        for image_name in self.images:
-            classification = self.annotations[image_name]
-            image = self.images[image_name]
+        for image_path in self.images:
+            classification = self.annotations[image_path]
+            image = self.images[image_path]
+            image_name = Path(image_path).name
             class_id = (
                 classification.class_id[0]
                 if classification.confidence is None
@@ -666,9 +667,9 @@ def from_folder_structure(cls, root_directory_path: str) -> ClassificationDatase
             class_id = classes.index(class_name)
 
             for image in os.listdir(os.path.join(root_directory_path, class_name)):
-                image_dir = os.path.join(root_directory_path, class_name, image)
-                images[image] = cv2.imread(image_dir)
-                annotations[image] = Classifications(
+                image_path = str(os.path.join(root_directory_path, class_name, image))
+                images[image_path] = cv2.imread(image_path)
+                annotations[image_path] = Classifications(
                     class_id=np.array([class_id]),
                 )
 

supervision/dataset/formats/coco.py

@@ -159,7 +159,7 @@ def load_coco_annotations(
         image_annotations = coco_annotations_groups.get(coco_image["id"], [])
         image_path = os.path.join(images_directory_path, image_name)
 
-        image = cv2.imread(str(image_path))
+        image = cv2.imread(image_path)
         annotation = coco_annotations_to_detections(
             image_annotations=image_annotations,
             resolution_wh=(image_width, image_height),
@@ -170,8 +170,8 @@ def load_coco_annotations(
             detections=annotation,
         )
 
-        images[image_name] = image
-        annotations[image_name] = annotation
+        images[image_path] = image
+        annotations[image_path] = annotation
 
     return classes, images, annotations
 
@@ -200,9 +200,9 @@ def save_coco_annotations(
     coco_categories = classes_to_coco_categories(classes=classes)
 
     image_id, annotation_id = 1, 1
-    for image_name, image in images.items():
+    for image_path, image in images.items():
         image_height, image_width, _ = image.shape
-
+        image_name = f"{Path(image_path).stem}{Path(image_path).suffix}"
         coco_image = {
             "id": image_id,
             "license": 1,
@@ -213,7 +213,7 @@ def save_coco_annotations(
         }
 
         coco_images.append(coco_image)
-        detections = annotations[image_name]
+        detections = annotations[image_path]
 
         coco_annotation, annotation_id = detections_to_coco_annotations(
             detections=detections,

supervision/dataset/formats/pascal_voc.py

@@ -168,12 +168,13 @@ def load_pascal_voc_annotations(
 
     for image_path in image_paths:
         image_name = Path(image_path).stem
-        image = cv2.imread(str(image_path))
+        image_path = str(image_path)
+        image = cv2.imread(image_path)
 
         annotation_path = os.path.join(annotations_directory_path, f"{image_name}.xml")
         if not os.path.exists(annotation_path):
-            images[image_path.name] = image
-            annotations[image_path.name] = Detections.empty()
+            images[image_path] = image
+            annotations[image_path] = Detections.empty()
             continue
 
         tree = parse(annotation_path)
@@ -184,8 +185,8 @@ def load_pascal_voc_annotations(
             root, classes, resolution_wh, force_masks
         )
 
-        images[image_path.name] = image
-        annotations[image_path.name] = annotation
+        images[image_path] = image
+        annotations[image_path] = annotation
 
     return classes, images, annotations
 

supervision/dataset/formats/yolo.py

@@ -139,13 +139,14 @@ def load_yolo_annotations(
     annotations = {}
 
     for image_path in image_paths:
-        image_name = Path(image_path).stem
-        image = cv2.imread(str(image_path))
+        image_stem = Path(image_path).stem
+        image_path = str(image_path)
+        image = cv2.imread(image_path)
 
-        annotation_path = os.path.join(annotations_directory_path, f"{image_name}.txt")
+        annotation_path = os.path.join(annotations_directory_path, f"{image_stem}.txt")
         if not os.path.exists(annotation_path):
-            images[image_path.name] = image
-            annotations[image_path.name] = Detections.empty()
+            images[image_path] = image
+            annotations[image_path] = Detections.empty()
             continue
 
         lines = read_txt_file(str(annotation_path))
@@ -158,8 +159,8 @@ def load_yolo_annotations(
             lines=lines, resolution_wh=resolution_wh, with_masks=with_masks
         )
 
-        images[image_path.name] = image
-        annotations[image_path.name] = annotation
+        images[image_path] = image
+        annotations[image_path] = annotation
     return classes, images, annotations
 
 
@@ -227,8 +228,9 @@ def save_yolo_annotations(
     approximation_percentage: float = 0.75,
 ) -> None:
     Path(annotations_directory_path).mkdir(parents=True, exist_ok=True)
-    for image_name, image in images.items():
-        detections = annotations[image_name]
+    for image_path, image in images.items():
+        detections = annotations[image_path]
+        image_name = Path(image_path).name
         yolo_annotations_name = _image_name_to_annotation_name(image_name=image_name)
         yolo_annotations_path = os.path.join(
             annotations_directory_path, yolo_annotations_name

supervision/dataset/utils.py

@@ -97,7 +97,8 @@ def save_dataset_images(
 ) -> None:
     Path(images_directory_path).mkdir(parents=True, exist_ok=True)
 
-    for image_name, image in images.items():
+    for image_path, image in images.items():
+        image_name = Path(image_path).name
         target_image_path = os.path.join(images_directory_path, image_name)
         cv2.imwrite(target_image_path, image)