Resize image and pad with color to achieve desired resolution while maintaining aspect ratio. Args: image: The image to resize a

Perform Non-Maximum Merging (NMM) on segmentation predictions. Args: predictions: A 2D array of object detection predictions in

Perform non-maximum merging on the current set of object detections. Dispatch order: (1) if mask data present, IoU mask is used; (2)

Generate *num_masks* random boolean masks with at least one True pixel each.

Split a flat F-order RLE into per-column run lists. With F-order (column-major) RLE the flat pixel sequence visits all rows of column 0, then

Annotates the given scene with polygons based on the provided detections. Args: scene: The image where polygons will be

Annotates the given scene with halos based on the provided detections. Args: scene: The image where the halo effect will

Helper function to create or append to a data dictionary with class names if available. Args: class_ids: Array of class IDs.

Approximates a given polygon by reducing a certain percentage of points. This function uses the Ramer-Douglas-Peucker algorithm to simplify

Close the CSV file.

Calculate the Mean Average Recall metric based on the stored predictions and ground-truth, at different IoU thresholds and maximum de

Checks if the binary mask contains holes (background pixels fully enclosed by foreground pixels). Args: mask: 2D binary mask whe

Converts OpenCV image into Pillow image, handling BGR -> RGB conversion. Args: image: OpenCV image (in BGR format). Returns

Convert Supervision Detections to a numpy tensor for metric computation. Args: detections: Detections/Targets in the format of sv.De

Converts images provided either as Pillow images or OpenCV images into OpenCV format. Args: images: Images to be converted

Offset the masks in an array by the specified (x, y) amount. Args: masks: A 3D array of binary masks corresponding to the

Build a Pascal VOC ``<object>`` XML element for one detection. Coordinates are converted to 1-indexed Pascal VOC convention before writing. T

Converts a polygon represented by a NumPy array into a bounding box. Args: polygon: A polygon represented by a NumPy array of shape

Determines if the detections are within the polygon zone. Anchor points are calculated from original (unclipped) detection boxes to

Converts a color input in multiple formats to a standardized BGR format. Args: color: The color input to be converted, w

Add data to the metric, without computing the result. Return the metric itself to allow method chaining.

Add new predictions and targets to the metric, but do not compute the result. Args: predictions: The predicted detection

Wrap `text` to the specified maximum line length, respecting existing newlines. Falls back to str() if `text` is not already a string. A

Decode a COCO base-48 string to raw (delta-encoded) integers. Implements the variable-length base-48 codec from the COCO API (pycocotools). E

Build per-image stats tuples and delegate to class-level computation. Each stats tuple is ``(matches, confidence, class_ids, true_class_ids)`

Build per-image stats tuples and delegate to class-level computation. Each stats tuple is ``(matches, confidence, class_ids, true_class_ids)`

Return a per-keypoint color list for a single instance.

Return the number of ``True`` pixels in a run-length encoded mask. Args: rle: int32 array of run lengths as produced by :func:`_mask_to_r

Applies a colored overlay to the scene outside of the detected regions. Args: scene: The image where masks will be drawn

Draws stroke-only covariance ellipse outlines around each keypoint. Args: scene: The image to annotate. ``ImageType`` ac

Draws radially-fading covariance ellipses around each keypoint. Args: scene: The image to annotate. ``ImageType`` accept

Checks if the binary mask contains multiple unconnected foreground segments. Args: mask: 2D binary mask where `True` indicates foreg

Convert normalized bounding box coordinates to absolute pixel coordinates. Multiplies each bounding box coordinate by image size and divides

Get the size category of a detections object. Args: detections: The detections object. metric_target: Determines whether box

Get image width and height as a tuple `(width, height)` for various image formats. Supports both `numpy.ndarray` images (with shape `(H, W,

Get a generator that yields the frames of the video. Args: source_path: The path of the video file. stride: Indicates the in

Check if the MD5 hash of a file matches the original hash. Note: MD5 is used here for file integrity checking (detecting corruption), no

Loads YOLO annotations and returns class names, images, and their corresponding detections. Args: images_directory_path: The

Converts a binary mask into a COCO run-length encoding (RLE). Produces RLE in the COCO format used by ``pycocotools``: pixels are counted

Merge metadata from a list of metadata dictionaries. This function combines the metadata dictionaries. If a key appears in more than one

Perform Non-Maximum Merging on oriented bounding box predictions, grouped per category. Mirrors :func:`box_non_max_merge` but uses orien

Pads bounding boxes coordinates with a constant padding. Args: xyxy: A numpy array of shape `(N, 4)` where each row corr

Run Kalman filter prediction step. Args: mean: The object state mean at the previous time step. covariance:

Re-encode all masks using tight bounding boxes. When the original ``xyxy`` boxes are padded or loose — common with object-detector ou

Converts a COCO run-length encoding (RLE) to a binary mask. Implements the COCO RLE format used by ``pycocotools``: pixels are counted i

Spread out boxes that overlap with each other. Args: xyxy: Numpy array of shape (N, 4) where N is the number of boxes. max_i

Convert bounding box to format `(center x, center y, aspect ratio, height)`, where the aspect ratio is `width / height`.

Convert oriented bounding box corners to axis-aligned bounding boxes. Args: xyxyxyxy: OBB corner coordinates with shape `(N, 4, 2)` where

Args: color: The color to use for the edges. thickness: The thickness of the edges. edges: The edges to d

Encode raw (delta-encoded) integers to a COCO base-48 string. The inverse of :func:`_base48_decode`. Applies the same variable-length base-48

Undo COCO delta encoding: ``counts[i] += counts[i - 2]`` for ``i > 2``. The COCO compressed RLE format stores run lengths as deltas relative to

Apply COCO delta encoding: ``d[i] = counts[i] - counts[i - 2]`` for ``i > 2``. The inverse of :func:`_delta_decode`. Converts absolute run length

Filter predictions and targets by object size category.

Filter predictions and targets by object size category.

Filter predictions and targets by object size category.

Filter predictions and targets by object size category.

Calculate the line counter angle (in degrees). Args: line_zone: The line zone object. Returns: Line

Return ellipse params grouped by sigma level (outermost first).

Internal function. Compute Intersection over Union (IoU) of two sets of masks - `masks_true` and `masks_detection`. Args:

Mux audio from `source_path` into `video_path` in-place using ffmpeg. Args: source_path: Path to the original video file containing the a

Transform targets into a dictionary that can be used by the COCO evaluator

Resize one RLE crop to ``(new_h, new_w)``, choosing the fastest path. Dispatch order: 1. **All-False fast path** — returns a single False ru

Return the label list for a single instance.

Concatenate per-column run lists into a flat RLE, merging junctions. Each column run list starts with a ``False``-run count. Two junction types

Run detection callback on a sliced portion of the image and adjust coordinates. Args: image: The full image.

Return the i-th element when the value stores per-detection data. Dispatch rules: - np.ndarray with ndim == 0: return as

Checks for shape consistency of input tensors.

Annotates the given scene by blurring regions based on the provided detections. Args: scene: The image where blurring wi

Returns the color as an RGB tuple. Returns: RGB tuple. Example: ```pycon >>> import sup

Exports the dataset to YOLO format. This method saves the images and their corresponding annotations in YOLO format. Args:

Perform Non-Maximum Suppression (NMS) on object detection predictions. Args: predictions: An array of object detection predictions i

Convert a list of class names to COCO ``categories`` entries. Category ids are emitted 1-indexed to comply with the COCO specification and to

Clips bounding boxes coordinates to fit within the frame resolution. Args: xyxy: A numpy array of shape `(N, 4)` where each

Convert COCO annotation dicts for a single image into a `Detections` object. .. warning:: The returned ``Detections.class_id`` contains *

Crop image based on bounding box coordinates. Args: image: The image to crop. xyxy: Bounding box coordinates in

Get array of cross products of each anchor with a vector. Args: anchors: Array of anchors of shape (number of anchors, detections, 2)

Converts an XML object in Pascal VOC format to a Detections object. Expected XML format: <annotation> ... <object>

Converts Detections object to Pascal VOC XML format. Args: detections: A Detections object containing bounding boxes, cl