🧹‍💨 clean up

setup.py

@@ -27,8 +27,7 @@ setuptools.setup(
         'numpy>=1.20.0',
         'opencv-python',
         'matplotlib',
-        'pyyaml',
-        'pillow'
+        'pyyaml'
     ],
     packages=find_packages(exclude=("tests",)),
     extras_require={

supervision/detection/annotate.py

@@ -140,94 +140,6 @@ class BoxAnnotator:
             )
         return scene
 
-    # Why a font file is needed?
-    # Non-ascii characters are not rendered properly with cv2.putText, so we need to use PIL
-    # PIL uses truetype fonts, so we need to provide a true type font file.
-
-    # Why not embed the font file in the package?
-    # The font file is not a part of the package because of licensing issues.
-
-    # Why not use a default font file?
-    # The default font file is not provided because of licensing issues. And the default font file may not be available on all systems or not suitable for all languages.
-
-    # Why a font size is needed?
-    # As for unicode support, different characters have different best visual sizes. So, we need to provide a font size to render the text properly.
-    def annotation_pil(
-        self,
-        scene: Image.Image,
-        detections: Detections,
-        font_file: str,
-        labels: Optional[List[str]] = None,
-        font_size: int = 15,
-        skip_label: bool = False,
-    ) -> Image.Image:
-        """
-        Draws bounding boxes on the frame using the detections provided with **PIL**. If non-ascii labels are provided, you should use this method.
-        For rendering non-ascii labels, you should provide a true type font file path to `font_file` argument.
-
-        Args:
-            scene (Image.Image): The image on which the bounding boxes will be drawn
-            detections (Detections): The detections for which the bounding boxes will be drawn
-            font_file (str): The true type font file path
-            font_size (int): The font size of the label.
-            labels (Optional[List[str]]): An optional list of labels corresponding to each detection. If `labels` are not provided, corresponding `class_id` will be used as label.
-        Returns:
-            Image.Image: The image with the bounding boxes drawn on it
-
-        Example:
-            ```python
-            >>> import supervision as sv
-            >>> from PIL import Image
-
-            >>> classes = ['person', ...]
-            >>> image = Image.open(...)
-            >>> font_file = Path('path/to/font/file').as_posix()
-            >>> detections = sv.Detections(...)
-
-            >>> box_annotator = sv.BoxAnnotator()
-            >>> labels = [
-            ...     '世界', '你好', 'hello', 'world'
-            ... ]
-            >>> annotated_frame = box_annotator.annotate_pil(
-            ...     scene=image,
-            ...     detections=detections,
-            ...     font_file=font_file,
-            ...     labels=labels
-            ... )
-            ```
-        """
-        color = ColorPalette.default()
-        padding_size = 3
-        text_color = "#fff"
-        draw = ImageDraw.Draw(scene)
-        font = ImageFont.truetype(font_file, font_size)
-        for i in range(len(detections)):
-            outline_color = color.by_idx(detections.class_id[i]).as_rgb()
-            x1, y1, x2, y2 = detections.xyxy[i].astype(int)
-            draw.rectangle((x1, y1, x2, y2), fill=None, outline=outline_color)
-            if skip_label:
-                continue
-            if labels:
-                text = str(labels[i])
-            else:
-                text = str(detections.class_id[i])
-            text_bbox = draw.textbbox((x1, y1), text, font=font)
-            text_height = text_bbox[3] - text_bbox[1]
-            text_x, text_y = x1, y1 - text_height - padding_size
-            text_bg_y = text_y + padding_size
-            draw.rectangle(
-                (
-                    text_x,
-                    text_bg_y,
-                    text_x + text_bbox[2] - text_bbox[0],
-                    text_bg_y + text_bbox[3] - text_bbox[1],
-                ),
-                fill=outline_color,
-            )
-            draw.text((text_x, text_y), text, font=font, fill=text_color)
-
-        return scene
-
 
 class MaskAnnotator:
     """