Update project version in pyproject.toml file

The project version in the pyproject.toml file has been updated from 0.17.0rc6 to 0.17.0rc7.

.github/requirements-dev-min.txt

@@ -0,0 +1,27 @@
+attrs==23.1.0
+certifi==2023.7.22
+charset-normalizer==2.0.12
+cycler==0.12.1
+exceptiongroup==1.1.3
+fonttools==4.43.1
+idna==3.4
+iniconfig==2.0.0
+kiwisolver==1.4.5
+matplotlib==3.5.0
+numpy==1.24.4
+opencv-python==4.5.5.64
+packaging==23.2
+Pillow==10.1.0
+pluggy==1.3.0
+pyparsing==3.1.1
+pytest==7.2.0
+python-dateutil==2.8.2
+PyYAML==5.3
+requests==2.26.0
+scipy==1.9.0
+setuptools-scm==8.0.4
+six==1.16.0
+tomli==2.0.1
+tqdm==4.62.3
+typing_extensions==4.8.0
+urllib3==1.26.18

.github/workflows/docs.yml

@@ -22,4 +22,5 @@ jobs:
           python-version: 3.x
       - run: pip install mkdocs-material
       - run: pip install "mkdocstrings[python]"
+      - run: pip install "mkdocs-material[imaging]"
       - run: mkdocs gh-deploy --force

.github/workflows/test-min.yml

@@ -0,0 +1,32 @@
+name: Python 3.8 - Min Dep Test WorkFlow
+
+on:
+  pull_request:
+    branches: [main, develop]
+
+jobs:
+  build-min-dep-test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.8"]
+    steps:
+      - name: 🛎️ Checkout
+        uses: actions/checkout@v4
+      - name: 🐍 Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v4
+        # id based on python version
+        id: python-setup
+        with:
+          python-version: ${{ matrix.python-version }}
+          check-latest: true
+          cache: 'pip'
+          cache-dependency-path: '**/requirements-dev-min.txt'
+
+      - name: 📦 Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -r .github/requirements-dev-min.txt
+
+      - name: 🧪 Test
+        run: "python -m pytest ./test"

.github/workflows/test.yml

@@ -5,7 +5,7 @@ on:
     branches: [main, develop]
 
 jobs:
-  build:
+  build-dev-test:
     runs-on: ubuntu-latest
     strategy:
       matrix:

.github/workflows/welcome.yml

@@ -11,7 +11,7 @@ jobs:
     name: 👋 Welcome
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/first-interaction@v1.2.0
+      - uses: actions/first-interaction@v1.3.0
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
           issue-message: "Hello there, thank you for opening an Issue ! 🙏🏻 The team was notified and they will get back to you asap."

.pre-commit-config.yaml

@@ -71,7 +71,7 @@ repos:
       -   id: black
 
   -   repo: https://github.com/astral-sh/ruff-pre-commit
-      rev: v0.1.5
+      rev: v0.1.6
       hooks:
       -   id: ruff
           args: [--fix, --exit-non-zero-on-fix]

README.md

@@ -41,66 +41,55 @@ Read more about desktop, headless, and local installation in our [guide](https:/
 
 ## 🔥 quickstart
 
-### [detections processing](https://roboflow.github.io/supervision/detection/core/)
+### models
+
+Supervision was designed to be model agnostic. Just plug in any classification, detection, or segmentation model. For your convenience, we have created [connectors](https://supervision.roboflow.com/detection/core/#detections) for the most popular libraries like Ultralytics, Transformers, or MMDetection.
 
 ```python
+>>> import cv2
 >>> import supervision as sv
 >>> from ultralytics import YOLO
 
+>>> image = cv2.imread(...)
 >>> model = YOLO('yolov8s.pt')
->>> result = model(IMAGE)[0]
+>>> result = model(image)[0]
 >>> detections = sv.Detections.from_ultralytics(result)
 
 >>> len(detections)
 5
 ```
 
-<details close>
-<summary>👉 more detections utils</summary>
+### annotators
 
-- Easily switch inference pipeline between supported object detection/instance segmentation models
+Supervision offers a wide range of highly customizable [annotators](https://supervision.roboflow.com/annotators/), allowing you to compose the perfect visualization for your use case.
 
-    ```python
-    >>> import supervision as sv
-    >>> from segment_anything import sam_model_registry, SamAutomaticMaskGenerator
+```python
+>>> import cv2
+>>> import supervision as sv
 
-    >>> sam = sam_model_registry[MODEL_TYPE](checkpoint=CHECKPOINT_PATH).to(device=DEVICE)
-    >>> mask_generator = SamAutomaticMaskGenerator(sam)
-    >>> sam_result = mask_generator.generate(IMAGE)
-    >>> detections = sv.Detections.from_sam(sam_result=sam_result)
-    ```
+>>> image = cv2.imread(...)
+>>> detections = sv.Detections(...)
 
-- [Advanced filtering](https://roboflow.github.io/supervision/quickstart/detections/)
+>>> bounding_box_annotator = sv.BoundingBoxAnnotator()
+>>> annotated_frame = bounding_box_annotator.annotate(
+...     scene=image.copy(),
+...     detections=detections
+... )
+```
 
-    ```python
-    >>> detections = detections[detections.class_id == 0]
-    >>> detections = detections[detections.confidence > 0.5]
-    >>> detections = detections[detections.area > 1000]
-    ```
+https://github.com/roboflow/supervision/assets/26109316/691e219c-0565-4403-9218-ab5644f39bce
 
-- Image annotation
+### datasets
 
-    ```python
-    >>> import supervision as sv
-
-    >>> box_annotator = sv.BoxAnnotator()
-    >>> annotated_frame = box_annotator.annotate(
-    ...     scene=IMAGE,
-    ...     detections=detections
-    ... )
-    ```
-
-</details>
-
-### [datasets processing](https://roboflow.github.io/supervision/dataset/core/)
+Supervision provides a set of [utils](https://supervision.roboflow.com/datasets/) that allow you to load, split, merge, and save datasets in one of the supported formats.
 
 ```python
 >>> import supervision as sv
 
 >>> dataset = sv.DetectionDataset.from_yolo(
-...     images_directory_path='...',
-...     annotations_directory_path='...',
-...     data_yaml_path='...'
+...     images_directory_path=...,
+...     annotations_directory_path=...,
+...     data_yaml_path=...
 ... )
 
 >>> dataset.classes
@@ -113,38 +102,27 @@ Read more about desktop, headless, and local installation in our [guide](https:/
 <details close>
 <summary>👉 more dataset utils</summary>
 
-- Load object detection/instance segmentation datasets in one of the supported formats
+- load
 
     ```python
     >>> dataset = sv.DetectionDataset.from_yolo(
-    ...     images_directory_path='...',
-    ...     annotations_directory_path='...',
-    ...     data_yaml_path='...'
+    ...     images_directory_path=...,
+    ...     annotations_directory_path=...,
+    ...     data_yaml_path=...
     ... )
 
     >>> dataset = sv.DetectionDataset.from_pascal_voc(
-    ...     images_directory_path='...',
-    ...     annotations_directory_path='...'
+    ...     images_directory_path=...,
+    ...     annotations_directory_path=...
     ... )
 
     >>> dataset = sv.DetectionDataset.from_coco(
-    ...     images_directory_path='...',
-    ...     annotations_path='...'
+    ...     images_directory_path=...,
+    ...     annotations_path=...
     ... )
     ```
 
-- Loop over dataset entries
-
-    ```python
-    >>> for name, image, labels in dataset:
-    ...     print(labels.xyxy)
-
-    array([[404.      , 719.      , 538.      , 884.5     ],
-           [155.      , 497.      , 404.      , 833.5     ],
-           [ 20.154999, 347.825   , 416.125   , 915.895   ]], dtype=float32)
-    ```
-
-- Split dataset for training, testing, and validation
+- split
 
     ```python
     >>> train_dataset, test_dataset = dataset.split(split_ratio=0.7)
@@ -154,7 +132,7 @@ Read more about desktop, headless, and local installation in our [guide](https:/
     (700, 150, 150)
     ```
 
-- Merge multiple datasets
+- merge
 
     ```python
     >>> ds_1 = sv.DetectionDataset(...)
@@ -176,103 +154,39 @@ Read more about desktop, headless, and local installation in our [guide](https:/
     ['cat', 'dog', 'person']
     ```
 
-- Save object detection/instance segmentation datasets in one of the supported formats
+- save
 
     ```python
     >>> dataset.as_yolo(
-    ...     images_directory_path='...',
-    ...     annotations_directory_path='...',
-    ...     data_yaml_path='...'
+    ...     images_directory_path=...,
+    ...     annotations_directory_path=...,
+    ...     data_yaml_path=...
     ... )
 
     >>> dataset.as_pascal_voc(
-    ...     images_directory_path='...',
-    ...     annotations_directory_path='...'
+    ...     images_directory_path=...,
+    ...     annotations_directory_path=...
     ... )
 
     >>> dataset.as_coco(
-    ...     images_directory_path='...',
-    ...     annotations_path='...'
+    ...     images_directory_path=...,
+    ...     annotations_path=...
     ... )
     ```
 
-- Convert labels between supported formats
+- convert
 
     ```python
     >>> sv.DetectionDataset.from_yolo(
-    ...     images_directory_path='...',
-    ...     annotations_directory_path='...',
-    ...     data_yaml_path='...'
+    ...     images_directory_path=...,
+    ...     annotations_directory_path=...,
+    ...     data_yaml_path=...
     ... ).as_pascal_voc(
-    ...     images_directory_path='...',
-    ...     annotations_directory_path='...'
+    ...     images_directory_path=...,
+    ...     annotations_directory_path=...
     ... )
     ```
 
-- Load classification datasets in one of the supported formats
-
-    ```python
-    >>> cs = sv.ClassificationDataset.from_folder_structure(
-    ...     root_directory_path='...'
-    ... )
-    ```
-
-- Save classification datasets in one of the supported formats
-
-    ```python
-    >>> cs.as_folder_structure(
-    ...     root_directory_path='...'
-    ... )
-    ```
-
-</details>
-
-### [model evaluation](https://roboflow.github.io/supervision/metrics/detection/)
-
-```python
->>> import supervision as sv
-
->>> dataset = sv.DetectionDataset.from_yolo(...)
-
->>> def callback(image: np.ndarray) -> sv.Detections:
-...     ...
-
->>> confusion_matrix = sv.ConfusionMatrix.benchmark(
-...     dataset = dataset,
-...     callback = callback
-... )
-
->>> confusion_matrix.matrix
-array([
-    [0., 0., 0., 0.],
-    [0., 1., 0., 1.],
-    [0., 1., 1., 0.],
-    [1., 1., 0., 0.]
-])
-```
-
-<details close>
-<summary>👉 more metrics</summary>
-
-- Mean average precision (mAP) for object detection tasks.
-
-    ```python
-    >>> import supervision as sv
-
-    >>> dataset = sv.DetectionDataset.from_yolo(...)
-
-    >>> def callback(image: np.ndarray) -> sv.Detections:
-    ...     ...
-
-    >>> mean_average_precision = sv.MeanAveragePrecision.benchmark(
-    ...     dataset = dataset,
-    ...     callback = callback
-    ... )
-
-    >>> mean_average_precision.map50_95
-    0.433
-    ```
-
 </details>
 
 ## 🎬 tutorials

docs/annotators.md

@@ -229,6 +229,27 @@
 
     </div>
 
+=== "Pixelate"
+
+    ```python
+    >>> import supervision as sv
+
+    >>> image = ...
+    >>> detections = sv.Detections(...)
+
+    >>> pixelate_annotator = sv.PixelateAnnotator()
+    >>> annotated_frame = pixelate_annotator.annotate(
+    ...     scene=image.copy(),
+    ...     detections=detections
+    ... )
+    ```
+
+    <div class="result" markdown>
+
+    ![pixelate-annotator-example](https://media.roboflow.com/supervision-annotator-examples/pixelate-annotator-example-10.png){ align=center width="800" }
+
+    </div>
+
 === "Trace"
 
     ```python
@@ -337,6 +358,10 @@
 
 :::supervision.annotators.core.BlurAnnotator
 
+## PixelateAnnotator
+
+:::supervision.annotators.core.PixelateAnnotator
+
 ## TraceAnnotator
 
 :::supervision.annotators.core.TraceAnnotator

docs/draw/utils.md

@@ -21,3 +21,11 @@
 ## draw_image
 
 :::supervision.draw.utils.draw_image
+
+## calculate_dynamic_font_scale
+
+:::supervision.draw.utils.calculate_dynamic_text_scale
+
+## calculate_dynamic_line_thickness
+
+:::supervision.draw.utils.calculate_dynamic_line_thickness

mkdocs.yml

@@ -1,7 +1,7 @@
 site_name: Supervision
 site_url: https://supervision.roboflow.com/
 site_author: Roboflow
-site_description: A set of easy-to-use utils that will come in handy in any Computer Vision project
+site_description: A set of easy-to-use utilities that will come in handy in any computer vision project.
 repo_name: roboflow/supervision
 repo_url: https://github.com/roboflow/supervision
 edit_uri: https://github.com/roboflow/supervision/tree/main/docs
@@ -60,8 +60,8 @@ nav:
 
 theme:
   name: 'material'
-  logo: https://media.roboflow.com/open-source/supervision/supervision-lenny.png
-  favicon: https://media.roboflow.com/open-source/supervision/supervision-lenny.png
+  logo: assets/supervision-lenny.png
+  favicon: assets/supervision-lenny.png
   palette:
     # Palette for light mode
     - scheme: default
@@ -79,10 +79,15 @@ theme:
   font:
     text: Roboto
     code: Roboto Mono
+  features:
+    - content.code.copy
 
 plugins:
   - mkdocstrings
   - search
+  - social:
+      cards_layout_dir: docs/layouts
+      cards_layout: custom
 
 markdown_extensions:
   - admonition

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "supervision"
-version = "0.17.0rc3"
+version = "0.17.0rc7"
 description = "A set of easy-to-use utils that will come in handy in any Computer Vision project"
 authors = ["Piotr Skalski <piotr.skalski92@gmail.com>"]
 maintainers = ["Piotr Skalski <piotr.skalski92@gmail.com>"]
@@ -36,15 +36,15 @@ classifiers=[
 
 [tool.poetry.dependencies]
 python = ">=3.8,<3.12.0"
-numpy = "^1.20.0"
-matplotlib = "^3.7.1"
-pyyaml = "^6.0"
-pillow = ">=9.4,<11.0"
-opencv-python = { version = "^4.8.0.74", optional = true }
-opencv-python-headless = "^4.8.0.74"
-scipy = "^1.9.0"
-requests = { version = "^2.31.0", optional = true }
-tqdm = { version = "^4.66.1", optional = true }
+numpy = ">=1.24.0,>=1.26.1"
+matplotlib = ">=3.5.0,>=3.8.1"
+pyyaml = ">=5.3,<=6.0.1"
+pillow = ">=9.4,<=10.1.0"
+opencv-python = { version = ">=4.5.5.64,<=4.8.1.78", optional = true }
+opencv-python-headless = ">=4.5.5.64,<=4.8.1.78"
+scipy = ">=1.9.0,<=2.0.0"
+requests = { version = ">=2.26.0,<=2.31.0", optional = true }
+tqdm = { version = ">=4.62.3,<=4.66.1", optional = true }
 
 
 [tool.poetry.extras]
@@ -57,7 +57,7 @@ pytest = "^7.2.2"
 wheel = ">=0.40,<0.42"
 notebook = ">=6.5.3,<8.0.0"
 build = ">=0.10,<1.1"
-ruff = ">=0.0.280,<0.1.6"
+ruff = ">=0.0.280,<0.1.7"
 isort = "^5.12.0"
 black = "^23.7.0"
 mypy = "^1.4.1"

supervision/__init__.py

@@ -18,6 +18,7 @@ from supervision.annotators.core import (
     HeatMapAnnotator,
     LabelAnnotator,
     MaskAnnotator,
+    PixelateAnnotator,
     PolygonAnnotator,
     TraceAnnotator,
 )
@@ -45,6 +46,8 @@ from supervision.detection.utils import (
 )
 from supervision.draw.color import Color, ColorPalette
 from supervision.draw.utils import (
+    calculate_dynamic_line_thickness,
+    calculate_dynamic_text_scale,
     draw_filled_rectangle,
     draw_image,
     draw_line,

supervision/annotators/core.py

@@ -30,7 +30,7 @@ class BoundingBoxAnnotator(BaseAnnotator):
                 annotating detections.
             thickness (int): Thickness of the bounding box lines.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.thickness: int = thickness
@@ -112,7 +112,7 @@ class MaskAnnotator(BaseAnnotator):
                 annotating detections.
             opacity (float): Opacity of the overlay mask. Must be between `0` and `1`.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.opacity = opacity
@@ -195,7 +195,7 @@ class PolygonAnnotator(BaseAnnotator):
                 annotating detections.
             thickness (int): Thickness of the polygon lines.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.thickness: int = thickness
@@ -277,7 +277,7 @@ class BoxMaskAnnotator(BaseAnnotator):
                 annotating detections.
             opacity (float): Opacity of the overlay mask. Must be between `0` and `1`.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.color_lookup: ColorLookup = color_lookup
@@ -366,7 +366,7 @@ class HaloAnnotator(BaseAnnotator):
             kernel_size (int): The size of the average pooling kernel used for creating
                 the halo.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.opacity = opacity
@@ -459,7 +459,7 @@ class EllipseAnnotator(BaseAnnotator):
             start_angle (int): Starting angle of the ellipse.
             end_angle (int): Ending angle of the ellipse.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.thickness: int = thickness
@@ -547,7 +547,7 @@ class BoxCornerAnnotator(BaseAnnotator):
             thickness (int): Thickness of the corner lines.
             corner_length (int): Length of each corner line.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.thickness: int = thickness
@@ -631,7 +631,7 @@ class CircleAnnotator(BaseAnnotator):
                 annotating detections.
             thickness (int): Thickness of the circle line.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
 
         self.color: Union[Color, ColorPalette] = color
@@ -717,7 +717,7 @@ class DotAnnotator(BaseAnnotator):
             radius (int): Radius of the drawn dots.
             position (Position): The anchor position for placing the dot.
             color_lookup (ColorLookup): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.radius: int = radius
@@ -800,7 +800,7 @@ class LabelAnnotator:
             text_position (Position): Position of the text relative to the detection.
                 Possible values are defined in the `Position` enum.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.text_color: Color = text_color
@@ -1027,7 +1027,7 @@ class TraceAnnotator:
                 points. Defaults to `30`.
             thickness (int): The thickness of the trace lines. Defaults to `2`.
             color_lookup (str): Strategy for mapping colors to annotations.
-                Options are `INDEX`, `CLASS`, `TRACE`.
+                Options are `INDEX`, `CLASS`, `TRACK`.
         """
         self.color: Union[Color, ColorPalette] = color
         self.position = position
@@ -1197,3 +1197,69 @@ class HeatMapAnnotator:
             mask
         ]
         return scene
+
+
+class PixelateAnnotator(BaseAnnotator):
+    """
+    A class for pixelating regions in an image using provided detections.
+    """
+
+    def __init__(self, pixel_size: int = 20):
+        """
+        Args:
+            pixel_size (int): The size of the pixelation.
+        """
+        self.pixel_size: int = pixel_size
+
+    def annotate(
+        self,
+        scene: np.ndarray,
+        detections: Detections,
+    ) -> np.ndarray:
+        """
+        Annotates the given scene by pixelating regions based on the provided
+            detections.
+
+        Args:
+            scene (np.ndarray): The image where pixelating will be applied.
+            detections (Detections): Object detections to annotate.
+
+        Returns:
+            The annotated image.
+
+        Example:
+            ```python
+            >>> import supervision as sv
+
+            >>> image = ...
+            >>> detections = sv.Detections(...)
+
+            >>> pixelate_annotator = sv.PixelateAnnotator()
+            >>> annotated_frame = pixelate_annotator.annotate(
+            ...     scene=image.copy(),
+            ...     detections=detections
+            ... )
+            ```
+
+        ![pixelate-annotator-example](https://media.roboflow.com/
+        supervision-annotator-examples/pixelate-annotator-example-10.png)
+        """
+        image_height, image_width = scene.shape[:2]
+        clipped_xyxy = clip_boxes(
+            xyxy=detections.xyxy, resolution_wh=(image_width, image_height)
+        ).astype(int)
+
+        for x1, y1, x2, y2 in clipped_xyxy:
+            roi = scene[y1:y2, x1:x2]
+            scaled_up_roi = cv2.resize(
+                src=roi, dsize=None, fx=1 / self.pixel_size, fy=1 / self.pixel_size
+            )
+            scaled_down_roi = cv2.resize(
+                src=scaled_up_roi,
+                dsize=(roi.shape[1], roi.shape[0]),
+                interpolation=cv2.INTER_NEAREST,
+            )
+
+            scene[y1:y2, x1:x2] = scaled_down_roi
+
+        return scene

supervision/classification/core.py

@@ -39,15 +39,58 @@ class Classifications:
         _validate_class_ids(self.class_id, n)
         _validate_confidence(self.confidence, n)
 
+    def __len__(self) -> int:
+        """
+        Returns the number of classifications.
+        """
+        return len(self.class_id)
+
+    @classmethod
+    def from_clip(cls, clip_results) -> Classifications:
+        """
+        Creates a Classifications instance from a
+        [clip](https://github.com/openai/clip) inference result.
+
+        Args:
+            clip_results (np.ndarray): The inference result from clip model.
+
+        Returns:
+            Classifications: A new Classifications object.
+
+        Example:
+            ```python
+            >>> from PIL import Image
+            >>> import clip
+            >>> import supervision as sv
+
+            >>> model, preprocess = clip.load('ViT-B/32')
+
+            >>> image = cv2.imread(SOURCE_IMAGE_PATH)
+            >>> image = preprocess(image).unsqueeze(0)
+
+            >>> text = clip.tokenize(["a diagram", "a dog", "a cat"])
+            >>> output, _ = model(image, text)
+            >>> classifications = sv.Classifications.from_clip(output)
+            ```
+        """
+
+        confidence = clip_results.softmax(dim=-1).cpu().detach().numpy()[0]
+
+        if len(confidence) == 0:
+            return cls(class_id=np.array([]), confidence=np.array([]))
+
+        class_ids = np.arange(len(confidence))
+        return cls(class_id=class_ids, confidence=confidence)
+
     @classmethod
     def from_ultralytics(cls, ultralytics_results) -> Classifications:
         """
         Creates a Classifications instance from a
-        (https://github.com/ultralytics/ultralytics) inference result.
+        [ultralytics](https://github.com/ultralytics/ultralytics) inference result.
 
         Args:
             ultralytics_results (ultralytics.engine.results.Results):
-                The output Results instance from ultralytics model
+                The inference result from ultralytics model.
 
         Returns:
             Classifications: A new Classifications object.
@@ -60,15 +103,57 @@ class Classifications:
 
             >>> image = cv2.imread(SOURCE_IMAGE_PATH)
             >>> model = YOLO('yolov8n-cls.pt')
-            >>> model = YOLO('yolov8s-cls.pt')
 
-            >>> result = model(image)[0]
-            >>> classifications = sv.Classifications.from_ultralytics(result)
+            >>> output = model(image)[0]
+            >>> classifications = sv.Classifications.from_ultralytics(output)
             ```
         """
         confidence = ultralytics_results.probs.data.cpu().numpy()
         return cls(class_id=np.arange(confidence.shape[0]), confidence=confidence)
 
+    @classmethod
+    def from_timm(cls, timm_results) -> Classifications:
+        """
+        Creates a Classifications instance from a
+        [timm](https://huggingface.co/docs/hub/timm) inference result.
+
+        Args:
+            timm_results: The inference result from timm model.
+
+        Returns:
+            Classifications: A new Classifications object.
+
+        Example:
+            ```python
+            >>> from PIL import Image
+            >>> import timm
+            >>> from timm.data import resolve_data_config, create_transform
+            >>> import supervision as sv
+
+            >>> model = timm.create_model(
+            ...     model_name='hf-hub:nateraw/resnet50-oxford-iiit-pet',
+            ...     pretrained=True
+            ... ).eval()
+
+            >>> config = resolve_data_config({}, model=model)
+            >>> transform = create_transform(**config)
+
+            >>> image = Image.open(SOURCE_IMAGE_PATH).convert('RGB')
+            >>> x = transform(image).unsqueeze(0)
+
+            >>> output = model(x)
+
+            >>> classifications = sv.Classifications.from_timm(output)
+            ```
+        """
+        confidence = timm_results.cpu().detach().numpy()[0]
+
+        if len(confidence) == 0:
+            return cls(class_id=np.array([]), confidence=np.array([]))
+
+        class_id = np.arange(len(confidence))
+        return cls(class_id=class_id, confidence=confidence)
+
     def get_top_k(self, k: int) -> Tuple[np.ndarray, np.ndarray]:
         """
         Retrieve the top k class IDs and confidences,

supervision/detection/core.py

@@ -324,7 +324,7 @@ class Detections:
 
             >>> inferencer = DetInferencer(model_name, checkpoint, device)
             >>> mmdet_result = inferencer(SOURCE_IMAGE_PATH, out_dir='./output',
-            ...                           return_datasample=True)["predictions"][0]
+            ...                           return_datasamples=True)["predictions"][0]
             >>> detections = sv.Detections.from_mmdet(mmdet_result)
             ```
         """

supervision/draw/color.py

@@ -71,6 +71,21 @@ class Color:
         r, g, b = (int(color_hex[i : i + 2], 16) for i in range(0, 6, 2))
         return cls(r, g, b)
 
+    def as_hex(self) -> str:
+        """
+        Converts the Color instance to a hex string.
+
+        Returns:
+            str: The hexadecimal color string.
+
+        Example:
+            ```
+            >>> Color(r=255, g=0, b=255).as_hex()
+            '#ff00ff'
+            ```
+        """
+        return f"#{self.r:02x}{self.g:02x}{self.b:02x}"
+
     def as_rgb(self) -> Tuple[int, int, int]:
         """
         Returns the color as an RGB tuple.

supervision/draw/utils.py

@@ -1,5 +1,5 @@
 import os
-from typing import Optional, Union
+from typing import Optional, Tuple, Union
 
 import cv2
 import numpy as np
@@ -231,3 +231,41 @@ def draw_image(
     scene[rect.y : rect.y + rect.height, rect.x : rect.x + rect.width] = blended_roi
 
     return scene
+
+
+def calculate_dynamic_text_scale(resolution_wh: Tuple[int, int]) -> float:
+    """
+    Calculate a dynamic font scale based on the resolution of an image.
+
+    Parameters:
+         resolution_wh (Tuple[int, int]): A tuple representing the width and height
+                 of the image.
+
+    Returns:
+         float: The calculated font scale factor.
+    """
+    return min(resolution_wh) * 1e-3
+
+
+def calculate_dynamic_line_thickness(resolution_wh: Tuple[int, int]) -> int:
+    """
+    Calculate a dynamic line thickness based on the resolution of an image.
+
+    Parameters:
+        resolution_wh (Tuple[int, int]): A tuple representing the width and height
+                of the image.
+
+    Returns:
+        int: The calculated line thickness in pixels.
+    """
+    min_dimension = min(resolution_wh)
+    if min_dimension < 480:
+        return 2
+    if min_dimension < 720:
+        return 2
+    if min_dimension < 1080:
+        return 2
+    if min_dimension < 2160:
+        return 4
+    else:
+        return 4

test/draw/test_color.py

@@ -30,3 +30,22 @@ def test_color_from_hex(
     with exception:
         result = Color.from_hex(color_hex=color_hex)
         assert result == expected_result
+
+
+@pytest.mark.parametrize(
+    "color, expected_result, exception",
+    [
+        (Color.white(), "#ffffff", DoesNotRaise()),
+        (Color.black(), "#000000", DoesNotRaise()),
+        (Color.red(), "#ff0000", DoesNotRaise()),
+        (Color.green(), "#00ff00", DoesNotRaise()),
+        (Color.blue(), "#0000ff", DoesNotRaise()),
+        (Color(r=128, g=128, b=0), "#808000", DoesNotRaise()),
+    ],
+)
+def test_color_as_hex(
+    color: Color, expected_result: Optional[str], exception: Exception
+) -> None:
+    with exception:
+        result = color.as_hex()
+        assert result == expected_result