The `sv.Detections` class in the Supervision library standardizes results from various object detection and segmentation models into a consis

F1 Score is a metric used to evaluate object detection models. It is the harmonic mean of precision and recall, calculated at different IoU t

Memory-efficient crop-RLE mask storage for instance segmentation. Instead of storing N full ``(H, W)`` boolean arrays, :class:`CompactMask` e

Precision is a metric used to evaluate object detection models. It is the ratio of true positive detections to the total number of predicted

Recall is a metric used to evaluate object detection models. It is the ratio of true positive detections to the total number of ground truth

Contains information about a detection dataset. Handles lazy image loading and annotation retrieval, dataset splitting, conversions into mult

Represents a point in 2D space. Attributes: x: The x-coordinate of the point. y: The y-coordinate of the point. Example

Mean Average Precision (mAP) is a metric used to evaluate object detection models. It is the average of the precision-recall curves at differ

Perform tiled inference on large images by slicing them into overlapping patches. This class divides an input image into overlapping slices

The `sv.KeyPoints` class in the Supervision library standardizes results from various keypoint detection and pose estimation models into a co

A class for drawing trace paths on an image based on detection coordinates. !!! warning This annotator uses the `sv.Detections.trac

This class is responsible for counting the number of objects that cross a predefined line. <video controls> <source

A class for pixelating regions in an image using provided detections.

A timer that calculates the duration each object has been detected based on the system clock. Attributes: tracker_id2start_time

Minimal tensor wrapper for cpu().numpy() and int().

A class for drawing Halos on an image using provided detections. !!! warning This annotator uses `sv.Detections.mask`.

A class for drawing masks on an image using provided detections. !!! warning This annotator uses `sv.Detections.mask`.

Mean Average Recall (mAR) measures how well the model detects and retrieves relevant objects by averaging recall over multiple IoU thresh

Represents a rectangle in 2D space. Attributes: x: The x-coordinate of the top-left corner of the rectangle. y: The y-coordi

A class for blurring regions in an image using provided detections.

A class for drawing bounding boxes on an image using provided detections.

A utility class for smoothing detections over multiple frames in video tracking. It maintains a history of detections for each track and prov

A class for drawing heatmaps on an image based on provided detections. Heat accumulates over time and is drawn as a semi-transparent overlay

The result of the Mean Average Precision calculation. Defaults to `0` when no detections or targets are present. Attributes: me

Represents a vector in 2D space, defined by a start and an end point. Attributes: start: The starting point of the vector. e

A class for drawing a colored overlay on the background of an image outside the region of detections. If masks are provided, the backgro

A timer that calculates the duration each object has been detected based on frames per second (FPS). Attributes: fps (float): Th

A class for drawing polygons on an image using provided detections. !!! warning This annotator uses `sv.Detections.mask`.

A class for drawing box corners on an image using provided detections.

A class for drawing circle on an image using provided detections.

Contains information about a classification dataset, handles lazy image loading, dataset splitting. Attributes: classes: List co

A class for drawing box masks on an image using provided detections.

Highlights the differences between two sets of detections. Useful for comparing results from two different models, or the difference betw

A class for drawing scaled up crops of detections on the scene.

A class for drawing dots on an image at specific coordinates based on provided detections.

A class for drawing ellipses on an image using provided detections.

The results of the F1 score metric calculation. Defaults to `0` if no detections or targets were provided. Attributes: metric_t

A simple Kalman filter for tracking bounding boxes in image space. The 8-dimensional state space is (x, y, a, h, vx, vy, va, vh), where

A class for annotating labels on an image using provided detections.

The results of the Mean Average Recall metric calculation. Defaults to `0` if no detections or targets were provided. Attributes:

A class for drawing oriented bounding boxes on an image using provided detections.

A class for drawing percentage bars on an image using provided detections.

The results of the precision metric calculation. Defaults to `0` if no detections or targets were provided. Attributes: metric_

The results of the recall metric calculation. Defaults to `0` if no detections or targets were provided. Attributes: metric_tar

A class for annotating labels on an image using provided detections, with support for Unicode characters by using a custom font.

A class for drawing bounding boxes with round edges on an image using provided detections.

A class for drawing triangle markers on an image at specific coordinates based on provided detections.

Ultralytics-like results container used by from_ultralytics.

Evaluator class to compute COCO metrics.

Parameters for COCOEvaluator

Represents a color in RGBA format. This class provides methods to work with colors, including creating colors from hex codes, converting

Class used representing a dataset in the right format needed by the `COCOEvaluator` class.

Enum specifying supported Large Multimodal Models (LMMs). .. deprecated:: 0.27.0 `LMM` is deprecated and will be removed in `supervi

Enum specifying supported Vision-Language Models (VLMs). Attributes: PALIGEMMA: Google's PaliGemma vision-language model. FL

A class to store video information, including width, height, fps and total number of frames. Attributes: width: width of the

Context manager that saves video frames to a file using OpenCV. Attributes: target_path: The path to the output file where the video

Ultralytics-like Boxes exposing xyxy/conf/cls and optional id.

YOLOv5-like results exposing pred list.

YOLO-NAS-like prediction struct.

YOLO-NAS-like results exposing prediction.

Defines different ways of averaging the metric results. Suppose, before returning the final result, a metric is computed for each class.

Initialize the ByteTrack object. .. deprecated:: 0.28.0 `ByteTrack` is deprecated since `supervision-0.28.0` and will be removed in

A utility class for saving detection data to a CSV file. This class is designed to efficiently serialize detection objects into a CSV format,

Enumeration class to define strategies for mapping colors to annotations. This enum supports three different lookup strategies: - `I

Confusion matrix for object detection tasks. Attributes: matrix: An 2D `np.ndarray` of shape `(len(classes) + 1, len(classes) + 1)`

A class that specializes in drawing skeleton edges on images using specified key points. It connects key points with lines to form the skelet

A class for monitoring frames per second (FPS) to benchmark latency.

A class for drawing an icon on an image, using provided detections.

Each member of this enum represents a image asset. The value associated with each member is the filename of the image. | Asset

A utility class for saving detection data to a JSON file. This class is designed to efficiently serialize detection objects into a JSON forma

!!! deprecated "Deprecated" `MeanAveragePrecision` is **deprecated** and will be removed in `supervision-0.31.0`. The de

The base class for all supervision metrics.

Specifies what type of detection is used to compute the metric. Attributes: BOXES: xyxy bounding boxes MASKS: Binary masks