roboflow/supervisionPublic

We write your reusable computer vision tools. 💜

AI summary: A reusable computer vision utility library for writing robust detection and tracking pipelines.

Stars
51.7K
+41 today
Forks
5K
Watchers
289
Open issues
46
Open PRs
58
Contributors
~249
Commits
5.5K
Branches
44

PythonMITCreated Nov 28, 2022Last push 1d agoLatest release 0.31.7+74 stars this week+427 this month

Quick answers

What is supervision?
A reusable computer vision utility library for writing robust detection and tracking pipelines.
What does supervision do?
Supervision is an open-source Python library designed to simplify the development of computer vision applications. It acts as a universal bridge between various object detection models (like YOLOv8, Ultralytics, and HuggingFace models) and practical application logic. Instead of writing boilerplate code for parsing complex model outputs, drawing bounding boxes, tracking object trajectories across frames, or filtering detections by specific zones, developers can use Supervision's high-level API. This dramatically accelerates the deployment of AI models into production environments by handling the tedious post-processing steps of computer vision workflows.
Who is supervision for?
Computer vision engineers, machine learning researchers, and developers building AI-powered video analytics applications. It is the go-to tool for rapidly translating raw model outputs into actionable data and visualizations.
How do I get started with supervision?
pip install supervision
How popular is supervision on GitHub?
roboflow/supervision has 51,676 stars and 4,958 forks on GitHub, and gained 74 stars in the last 7 days.
What license does supervision use?
roboflow/supervision is released under the MIT license.

Star history

since Jan 15, 2023
020K40KJan 2023Apr 2024Jul 2025Nov 2026
51.7K stars as of Nov 20, 2026. Before Jul 29, 2026, reconstructed from public GitHub event archives (checked against the repository's real star total); since then measured daily.

Update history

4 recorded
  • Nov 16, 2026New release 0.31.7 (previously 0.31.3).
  • Nov 1, 2026New release 0.31.3 (previously 0.31.1).
  • Oct 20, 2026New release 0.31.1 (previously 0.30.9).
  • Oct 9, 2026New release 0.30.9 (previously 0.30.5).

Contribution activity

commits per day, last 52 weeks
NovDecJanFebMarAprMayJunJulAugSepOctNovMonWedFri2025-11-23: 0 commits2025-11-24: 0 commits2025-11-25: 0 commits2025-11-26: 0 commits2025-11-27: 0 commits2025-11-28: 0 commits2025-11-29: 0 commits2025-11-30: 0 commits2025-12-01: 0 commits2025-12-02: 0 commits2025-12-03: 0 commits2025-12-04: 0 commits2025-12-05: 0 commits2025-12-06: 0 commits2025-12-07: 0 commits2025-12-08: 0 commits2025-12-09: 0 commits2025-12-10: 0 commits2025-12-11: 0 commits2025-12-12: 0 commits2025-12-13: 0 commits2025-12-14: 0 commits2025-12-15: 0 commits2025-12-16: 0 commits2025-12-17: 0 commits2025-12-18: 0 commits2025-12-19: 0 commits2025-12-20: 0 commits2025-12-21: 0 commits2025-12-22: 0 commits2025-12-23: 0 commits2025-12-24: 0 commits2025-12-25: 0 commits2025-12-26: 0 commits2025-12-27: 0 commits2025-12-28: 0 commits2025-12-29: 0 commits2025-12-30: 0 commits2025-12-31: 0 commits2026-01-01: 0 commits2026-01-02: 0 commits2026-01-03: 0 commits2026-01-04: 0 commits2026-01-05: 7 commits2026-01-06: 2 commits2026-01-07: 0 commits2026-01-08: 0 commits2026-01-09: 0 commits2026-01-10: 0 commits2026-01-11: 0 commits2026-01-12: 23 commits2026-01-13: 3 commits2026-01-14: 0 commits2026-01-15: 0 commits2026-01-16: 1 commit2026-01-17: 0 commits2026-01-18: 0 commits2026-01-19: 5 commits2026-01-20: 1 commit2026-01-21: 0 commits2026-01-22: 1 commit2026-01-23: 1 commit2026-01-24: 0 commits2026-01-25: 0 commits2026-01-26: 3 commits2026-01-27: 15 commits2026-01-28: 1 commit2026-01-29: 2 commits2026-01-30: 7 commits2026-01-31: 0 commits2026-02-01: 0 commits2026-02-02: 36 commits2026-02-03: 23 commits2026-02-04: 11 commits2026-02-05: 1 commit2026-02-06: 1 commit2026-02-07: 0 commits2026-02-08: 0 commits2026-02-09: 0 commits2026-02-10: 9 commits2026-02-11: 0 commits2026-02-12: 7 commits2026-02-13: 1 commit2026-02-14: 0 commits2026-02-15: 0 commits2026-02-16: 0 commits2026-02-17: 0 commits2026-02-18: 0 commits2026-02-19: 3 commits2026-02-20: 6 commits2026-02-21: 0 commits2026-02-22: 1 commit2026-02-23: 0 commits2026-02-24: 0 commits2026-02-25: 2 commits2026-02-26: 0 commits2026-02-27: 0 commits2026-02-28: 0 commits2026-03-01: 0 commits2026-03-02: 1 commit2026-03-03: 1 commit2026-03-04: 0 commits2026-03-05: 0 commits2026-03-06: 0 commits2026-03-07: 0 commits2026-03-08: 0 commits2026-03-09: 0 commits2026-03-10: 10 commits2026-03-11: 1 commit2026-03-12: 0 commits2026-03-13: 4 commits2026-03-14: 1 commit2026-03-15: 0 commits2026-03-16: 0 commits2026-03-17: 0 commits2026-03-18: 0 commits2026-03-19: 2 commits2026-03-20: 0 commits2026-03-21: 0 commits2026-03-22: 0 commits2026-03-23: 0 commits2026-03-24: 2 commits2026-03-25: 0 commits2026-03-26: 0 commits2026-03-27: 0 commits2026-03-28: 0 commits2026-03-29: 2 commits2026-03-30: 7 commits2026-03-31: 3 commits2026-04-01: 0 commits2026-04-02: 0 commits2026-04-03: 0 commits2026-04-04: 0 commits2026-04-05: 0 commits2026-04-06: 0 commits2026-04-07: 0 commits2026-04-08: 1 commit2026-04-09: 0 commits2026-04-10: 0 commits2026-04-11: 0 commits2026-04-12: 0 commits2026-04-13: 6 commits2026-04-14: 4 commits2026-04-15: 7 commits2026-04-16: 1 commit2026-04-17: 3 commits2026-04-18: 3 commits2026-04-19: 0 commits2026-04-20: 1 commit2026-04-21: 2 commits2026-04-22: 6 commits2026-04-23: 1 commit2026-04-24: 2 commits2026-04-25: 1 commit2026-04-26: 0 commits2026-04-27: 1 commit2026-04-28: 1 commit2026-04-29: 0 commits2026-04-30: 4 commits2026-05-01: 1 commit2026-05-02: 0 commits2026-05-03: 0 commits2026-05-04: 0 commits2026-05-05: 0 commits2026-05-06: 0 commits2026-05-07: 0 commits2026-05-08: 0 commits2026-05-09: 0 commits2026-05-10: 0 commits2026-05-11: 0 commits2026-05-12: 0 commits2026-05-13: 3 commits2026-05-14: 1 commit2026-05-15: 0 commits2026-05-16: 0 commits2026-05-17: 0 commits2026-05-18: 0 commits2026-05-19: 4 commits2026-05-20: 3 commits2026-05-21: 0 commits2026-05-22: 7 commits2026-05-23: 0 commits2026-05-24: 0 commits2026-05-25: 1 commit2026-05-26: 1 commit2026-05-27: 2 commits2026-05-28: 1 commit2026-05-29: 1 commit2026-05-30: 0 commits2026-05-31: 0 commits2026-06-01: 1 commit2026-06-02: 1 commit2026-06-03: 0 commits2026-06-04: 3 commits2026-06-05: 0 commits2026-06-06: 1 commit2026-06-07: 2 commits2026-06-08: 3 commits2026-06-09: 4 commits2026-06-10: 4 commits2026-06-11: 2 commits2026-06-12: 0 commits2026-06-13: 0 commits2026-06-14: 0 commits2026-06-15: 9 commits2026-06-16: 3 commits2026-06-17: 8 commits2026-06-18: 7 commits2026-06-19: 1 commit2026-06-20: 0 commits2026-06-21: 0 commits2026-06-22: 2 commits2026-06-23: 3 commits2026-06-24: 1 commit2026-06-25: 6 commits2026-06-26: 4 commits2026-06-27: 4 commits2026-06-28: 0 commits2026-06-29: 8 commits2026-06-30: 1 commit2026-07-01: 12 commits2026-07-02: 3 commits2026-07-03: 8 commits2026-07-04: 0 commits2026-07-05: 0 commits2026-07-06: 6 commits2026-07-07: 4 commits2026-07-08: 3 commits2026-07-09: 4 commits2026-07-10: 1 commit2026-07-11: 3 commits2026-07-12: 0 commits2026-07-13: 0 commits2026-07-14: 1 commit2026-07-15: 4 commits2026-07-16: 4 commits2026-07-17: 8 commits2026-07-18: 0 commits2026-07-19: 0 commits2026-07-20: 5 commits2026-07-21: 7 commits2026-07-22: 0 commits2026-07-23: 0 commits2026-07-24: 0 commits2026-07-25: 0 commits2026-07-26: 1 commit2026-07-27: 3 commits2026-07-28: 4 commits2026-07-29: 0 commits2026-07-30: 0 commits2026-07-31: 1 commit2026-08-01: 0 commits2026-08-02: 0 commits2026-08-03: 5 commits2026-08-04: 5 commits2026-08-05: 2 commits2026-08-06: 1 commit2026-08-07: 1 commit2026-08-08: 0 commits2026-08-09: 0 commits2026-08-10: 4 commits2026-08-11: 1 commit2026-08-12: 1 commit2026-08-13: 0 commits2026-08-14: 1 commit2026-08-15: 0 commits2026-08-16: 0 commits2026-08-17: 1 commit2026-08-18: 0 commits2026-08-19: 2 commits2026-08-20: 3 commits2026-08-21: 3 commits2026-08-22: 1 commit2026-08-23: 0 commits2026-08-24: 9 commits2026-08-25: 2 commits2026-08-26: 0 commits2026-08-27: 0 commits2026-08-28: 0 commits2026-08-29: 0 commits2026-08-30: 0 commits2026-08-31: 0 commits2026-09-01: 0 commits2026-09-02: 8 commits2026-09-03: 8 commits2026-09-04: 3 commits2026-09-05: 1 commit2026-09-06: 0 commits2026-09-07: 6 commits2026-09-08: 4 commits2026-09-09: 6 commits2026-09-10: 2 commits2026-09-11: 5 commits2026-09-12: 0 commits2026-09-13: 0 commits2026-09-14: 11 commits2026-09-15: 6 commits2026-09-16: 12 commits2026-09-17: 7 commits2026-09-18: 7 commits2026-09-19: 0 commits2026-09-20: 0 commits2026-09-21: 4 commits2026-09-22: 11 commits2026-09-23: 0 commits2026-09-24: 0 commits2026-09-25: 6 commits2026-09-26: 0 commits2026-09-27: 0 commits2026-09-28: 5 commits2026-09-29: 6 commits2026-09-30: 2 commits2026-10-01: 4 commits2026-10-02: 0 commits2026-10-03: 2 commits2026-10-04: 1 commit2026-10-05: 2 commits2026-10-06: 13 commits2026-10-07: 9 commits2026-10-08: 7 commits2026-10-09: 7 commits2026-10-10: 1 commit2026-10-11: 2 commits2026-10-12: 6 commits2026-10-13: 4 commits2026-10-14: 1 commit2026-10-15: 8 commits2026-10-16: 13 commits2026-10-17: 3 commits2026-10-18: 0 commits2026-10-19: 5 commits2026-10-20: 6 commits2026-10-21: 3 commits2026-10-22: 8 commits2026-10-23: 9 commits2026-10-24: 3 commits2026-10-25: 4 commits2026-10-26: 16 commits2026-10-27: 17 commits2026-10-28: 5 commits2026-10-29: 0 commits2026-10-30: 1 commit2026-10-31: 0 commits2026-11-01: 1 commit2026-11-02: 4 commits2026-11-03: 12 commits2026-11-04: 11 commits2026-11-05: 9 commits2026-11-06: 1 commit2026-11-07: 0 commits2026-11-08: 0 commits2026-11-09: 14 commits2026-11-10: 19 commits2026-11-11: 12 commits2026-11-12: 9 commits2026-11-13: 7 commits2026-11-14: 0 commits2026-11-15: 0 commits2026-11-16: 0 commits2026-11-17: 0 commits2026-11-18: 0 commits2026-11-19: 0 commits2026-11-20: 0 commits2026-11-21: 0 commits
857 commits in the last yearLessMore

Signals and awards

derived from tracked data
  • Landmark project

    51,676 stars

  • Very active

    857 commits in 52 weeks

  • Community-driven

    ~249 contributors

  • Permissive license

    MIT

  • Continuous integration

    Automated checks passing

  • Repeat trending

    8 trending appearances

What supervision does

Supervision is an open-source Python library designed to simplify the development of computer vision applications. It acts as a universal bridge between various object detection models (like YOLOv8, Ultralytics, and HuggingFace models) and practical application logic. Instead of writing boilerplate code for parsing complex model outputs, drawing bounding boxes, tracking object trajectories across frames, or filtering detections by specific zones, developers can use Supervision's high-level API. This dramatically accelerates the deployment of AI models into production environments by handling the tedious post-processing steps of computer vision workflows.

Computer vision engineers, machine learning researchers, and developers building AI-powered video analytics applications. It is the go-to tool for rapidly translating raw model outputs into actionable data and visualizations.

  • Model Agnostic API: Provides a unified interface to process outputs from almost any object detection or segmentation model.
  • Advanced Annotators: Includes a wide array of highly customizable drawing tools for bounding boxes, polygons, and labels.
  • Object Tracking Integration: Seamlessly integrates robust multi-object tracking algorithms to follow entities across video frames.
  • Zone Filtering: Allows developers to define specific polygonal zones to filter detections or trigger events based on spatial location.
  • Dataset Management: Offers utilities for converting, filtering, and merging computer vision datasets in various formats.

Where teams use it

Video Analytics

Build a pipeline to count vehicles crossing a specific line in a traffic camera feed.

Retail Monitoring

Track customer movement and dwell times within specific zones of a retail store using security footage.

Model Evaluation

Visually debugging and evaluating the accuracy of a newly trained object detection model by plotting its predictions.

Manufacturing Inspection

Set up a rapid prototype to detect defects on a conveyor belt, filtering out objects outside the inspection area.

Getting started: pip install supervision

README

develop branch
📑 Table of Contents

👋 Hello

We are your essential toolkit for computer vision. From data loading to real-time zone counting, we provide the building blocks so you can focus on building applications around your models. 🤝

💻 Install

Pip install the supervision package in a Python>=3.10 environment.

pip install supervision

Read more about conda, mamba, and installing from source in our guide.

🔥 Quickstart

Models

Supervision was designed to be model agnostic. Just plug in any classification, detection, or segmentation model. For your convenience, we have created connectors for the most popular libraries like Ultralytics, Transformers, MMDetection, or Inference. Other integrations, like rfdetr, already return sv.Detections directly.

Install the optional dependencies for this example with pip install pillow rfdetr.

import supervision as sv
from PIL import Image
from rfdetr import RFDETRSmall

image = Image.open("path/to/image.jpg")
model = RFDETRSmall()
detections = model.predict(image, threshold=0.5)

len(detections)
# 5
👉 more model connectors
  • inference

    Running with Inference requires a Roboflow API KEY.

    import supervision as sv
    from PIL import Image
    from inference import get_model
    
    image = Image.open("path/to/image.jpg")
    model = get_model(model_id="rfdetr-small", api_key="ROBOFLOW_API_KEY")
    result = model.infer(image)[0]
    detections = sv.Detections.from_inference(result)
    
    len(detections)
    # 5

Annotators

Supervision offers a wide range of highly customizable annotators, allowing you to compose the perfect visualization for your use case.

import cv2
import supervision as sv

image = cv2.imread("path/to/image.jpg")
# Assuming detections are obtained from a model
detections = sv.Detections(...)

box_annotator = sv.BoxAnnotator()
annotated_frame = box_annotator.annotate(scene=image.copy(), detections=detections)
supervision-0.16.0-annotators.mp4

Datasets

Supervision provides a set of utils that allow you to load, split, merge, and save datasets in one of the supported formats.

import supervision as sv
from roboflow import Roboflow

project = Roboflow().workspace("WORKSPACE_ID").project("PROJECT_ID")
dataset = project.version("PROJECT_VERSION").download("coco")

ds = sv.DetectionDataset.from_coco(
    images_directory_path=f"{dataset.location}/train",
    annotations_path=f"{dataset.location}/train/_annotations.coco.json",
)

path, image, annotation = ds[0]
# loads image on demand

for path, image, annotation in ds:
    # loads image on demand
    pass
👉 more dataset utils
  • load

    dataset = sv.DetectionDataset.from_yolo(
        images_directory_path=...,
        annotations_directory_path=...,
        data_yaml_path=...,
    )
    
    dataset = sv.DetectionDataset.from_pascal_voc(
        images_directory_path=...,
        annotations_directory_path=...,
    )
    
    dataset = sv.DetectionDataset.from_coco(
        images_directory_path=...,
        annotations_path=...,
    )
  • split

    train_dataset, test_dataset = dataset.split(split_ratio=0.7)
    test_dataset, valid_dataset = test_dataset.split(split_ratio=0.5)
    
    len(train_dataset), len(test_dataset), len(valid_dataset)
    # (700, 150, 150)
  • merge

    ds_1 = sv.DetectionDataset(...)
    len(ds_1)
    # 100
    ds_1.classes
    # ['dog', 'person']
    
    ds_2 = sv.DetectionDataset(...)
    len(ds_2)
    # 200
    ds_2.classes
    # ['cat']
    
    ds_merged = sv.DetectionDataset.merge([ds_1, ds_2])
    len(ds_merged)
    # 300
    ds_merged.classes
    # ['cat', 'dog', 'person']
  • save

    dataset.as_yolo(
        images_directory_path=...,
        annotations_directory_path=...,
        data_yaml_path=...,
    )
    
    dataset.as_pascal_voc(
        images_directory_path=...,
        annotations_directory_path=...,
    )
    
    dataset.as_coco(
        images_directory_path=...,
        annotations_path=...,
    )
  • convert

    sv.DetectionDataset.from_yolo(
        images_directory_path=...,
        annotations_directory_path=...,
        data_yaml_path=...,
    ).as_pascal_voc(
        images_directory_path=...,
        annotations_directory_path=...,
    )

🎬 Tutorials

Want to learn how to use Supervision? Explore our how-to guides, end-to-end examples, cheatsheet, and cookbooks!


Dwell Time Analysis with Computer Vision | Real-Time Stream Processing Dwell Time Analysis with Computer Vision | Real-Time Stream Processing

Created: 5 Apr 2024

Learn how to use computer vision to analyze wait times and optimize processes. This tutorial covers object detection, tracking, and calculating time spent in designated zones. Use these techniques to improve customer experience in retail, traffic management, or other scenarios.


Speed Estimation & Vehicle Tracking | Computer Vision | Open Source Speed Estimation & Vehicle Tracking | Computer Vision | Open Source

Created: 11 Jan 2024

Learn how to track and estimate the speed of vehicles using YOLO, ByteTrack, and Roboflow Inference. This comprehensive tutorial covers object detection, multi-object tracking, filtering detections, perspective transformation, speed estimation, visualization improvements, and more.

💜 Built with Supervision

Did you build something cool using supervision? Let us know!

football-players-tracking-25.mp4
traffic_analysis_result.mov
vehicles-step-7-new.mp4

📚 Documentation

Visit our documentation page to learn how supervision can help you build computer vision applications faster and more reliably.

🏆 Contribution

We love your input! Please see our contributing guide to get started. Thank you 🙏 to all our contributors!


View on GitHub

Recent activity

commits and pull requests

Recent open issues

view all

Discussions

all 272

Releases and announcements

57 total
  1. supervision-0.31.70.31.7Nov 11, 202637 downloads

    # v0.31.7: pycocotools-matching mAP, stricter key point and CSV inputs, and edge-case fixes for zones and dataset exports A fixes-only patch: mAP now matches pycocotools more closely, and several zone, key point and export edge cases stop failing. - **MeanAveragePrecision** uses pycocotools size buckets and the COCO crowd IoU for oriented-box crowd targets. - **PolygonZone** keeps its full mask when polygon coordinates reach their integer limit. - Key point selection and **CSVSink** now raise on inputs they used to silently misread. - **DetectionDataset** exports keep box area for empty COCO masks and, with simplification on, bridge more mask holes. - The OpenCV-free `contourArea` and `convertScaleAbs` fallbacks agree with OpenCV in more cases. Drop-in upgrade, but check the Migration guide if you compare mAP values, pass extra CSV fields or select key points with boolean masks. ## ✨ Spotlights / highlights ### `MeanAveragePrecision` matches `pycocotools` more closely `sv.metrics.MeanAveragePrecision` now puts objects in the small, medium and large buckets that `pycocotools` gives them: a prediction's own geometry (box, mask or oriented-box polygon) decides, a

  2. supervision-0.31.60.31.6Nov 10, 202627 downloads

    # v0.31.6: Key points on empty frames, zero-shot Transformers results, and drawing on grayscale and palette scenes A fixes-only patch: fewer crashes on empty frames, more Transformers models that load, and annotators that draw on any scene. - **KeyPoints** filters stop raising `TypeError` on video frames with no people. - Grounding DINO and RF-DETR segmentation results load, and OWLv2 detections get real class names. - `DetectionsSmoother` keeps unconfirmed tracks as separate boxes instead of one averaged box. - Labels and images draw on grayscale, BGRA and Pillow palette scenes. - Integer boxes no longer wrap around, and narrow-integer mask shifts no longer crash. It's a drop-in upgrade, but check the Migration guide if you use ViTPose class IDs, YOLO-NAS confidences or compare stored outputs. ## ✨ Spotlights / highlights ### Key points survive empty frames `sv.KeyPoints.empty()` now holds empty `class_id`, `keypoint_confidence` and `detection_confidence` arrays instead of `None`, and selecting key points from it returns an empty result. Pose connectors such as `sv.KeyPoints.from_ultralytics` return it for every frame without people, so the documented filters

  3. supervision-0.31.50.31.5Nov 5, 202682 downloads

    # v0.31.5: Line counts through occlusion, the exact clipper for oriented boxes, and right-to-left labels Right-to-left labels, line counting that survives occlusion, and a batch of edge-case fixes across annotators, exports and the NumPy fallback. - **LineZone** counts objects that are hidden for a few frames while they cross the line. - Arabic and Hebrew labels render correctly with the new `rtl` extra. - Oriented-box IoU, NMS and metrics now use the exact clipper on OpenCV installs. - Annotating a Pillow image keeps its transparency and untouched pixels. - Heat maps and ellipse halos now work on grayscale scenes. It's a drop-in upgrade, but check the Migration guide if you count line crossings, score-oriented boxes on OpenCV, or compare stored outputs. ## ✨ Spotlights / highlights ### Right-to-left labels with the new `rtl` extra `sv.RichLabelAnnotator` now draws Arabic and Hebrew labels correctly on RGB and RGBA images when the new `rtl` extra is installed. Pillow only joins Arabic letters and orders right-to-left text when it's built with Raqm, and most Windows installs and many containers lack it. There, `"سيارة 0.92"` came out as isolated letters in reve

  4. supervision-0.31.40.31.4Nov 3, 202625 downloads

    # v0.31.4 — Stricter dataset export, COCO-exact recall buckets, and a closer NumPy fallback Fourteen bug fixes: stricter dataset export, COCO-exact recall buckets, sturdier VLM parsing, and a NumPy fallback closer to OpenCV. - **CreateML and Pascal VOC export** raise a clear `ValueError` for a `class_id` outside `classes`. - `DetectionDataset.merge` keeps annotations that have no class IDs instead of raising. - **MeanAverageRecall** counts boundary-area objects in both COCO size buckets, like `MeanAveragePrecision`. - Qwen and PaliGemma parsing keep valid boxes and class labels beside malformed ones. - Without OpenCV, image reads, gray conversion, drawing colors and keyword parameters match OpenCV. Drop-in upgrade; see the Migration guide if you export out-of-range class IDs or compare mAR values or fallback pixels against stored references. ## ✨ Spotlights / highlights ### Dataset export rejects out-of-range class IDs `detections_to_createml_annotations` and `detections_to_pascal_voc` now raise `ValueError` for a `class_id` outside `classes`, as `detections_to_labelme_shapes` already does. A negative id was exported under the last class name, and an id past t

  5. supervision-0.31.30.31.3Oct 28, 202630 downloads

    # v0.31.3 — Grayscale annotators, OpenCV-exact fallbacks, and merged YOLO polygons Grayscale scenes work with more annotators, the pure NumPy fallback matches OpenCV pixel for pixel, and YOLO export handles multi-part masks. - **YOLO export** merges multi-part masks into one polygon and skips crowd annotations, with flags for the old output. - Background overlay and comparison annotators draw on grayscale scenes. - Without OpenCV, resize, blur, blending, drawing and connected components give the same pixels as OpenCV. - Invalid input fails early: `ImageSink`, the video frame generator and depth arrays now raise clear errors. - Short hex colors such as `"#f0f"` work in annotators. Drop-in upgrade unless you export YOLO labels: the default output changes, see the Migration guide. ## ✨ Spotlights / highlights ### YOLO export merges multi-part masks and skips crowd annotations `DetectionDataset.as_yolo` now omits detections whose `iscrowd` flag is set and writes a mask with several disconnected parts as one merged polygon line instead of one line per part. Both are on by default. The merge is done by the new `sv.merge_multi_segment`, which joins the parts through

Code frequency

additions and deletions
+50.3K-50.3KWeek of 2025-11-23: +0 linesWeek of 2025-11-23: -0 linesWeek of 2025-11-30: +0 linesWeek of 2025-11-30: -0 linesWeek of 2025-12-07: +0 linesWeek of 2025-12-07: -0 linesWeek of 2025-12-14: +0 linesWeek of 2025-12-14: -0 linesWeek of 2025-12-21: +0 linesWeek of 2025-12-21: -0 linesWeek of 2025-12-28: +0 linesWeek of 2025-12-28: -0 linesWeek of 2026-01-04: +65 linesWeek of 2026-01-04: -29 linesWeek of 2026-01-11: +1,051 linesWeek of 2026-01-11: -891 linesWeek of 2026-01-18: +562 linesWeek of 2026-01-18: -86 linesWeek of 2026-01-25: +2,942 linesWeek of 2026-01-25: -902 linesWeek of 2026-02-01: +50,303 linesWeek of 2026-02-01: -46,494 linesWeek of 2026-02-08: +3,409 linesWeek of 2026-02-08: -1,202 linesWeek of 2026-02-15: +1,055 linesWeek of 2026-02-15: -137 linesWeek of 2026-02-22: +115 linesWeek of 2026-02-22: -4 linesWeek of 2026-03-01: +9 linesWeek of 2026-03-01: -9 linesWeek of 2026-03-08: +1,349 linesWeek of 2026-03-08: -1,079 linesWeek of 2026-03-15: +6 linesWeek of 2026-03-15: -7 linesWeek of 2026-03-22: +13 linesWeek of 2026-03-22: -13 linesWeek of 2026-03-29: +1,434 linesWeek of 2026-03-29: -75 linesWeek of 2026-04-05: +3 linesWeek of 2026-04-05: -3 linesWeek of 2026-04-12: +6,773 linesWeek of 2026-04-12: -686 linesWeek of 2026-04-19: +3,458 linesWeek of 2026-04-19: -334 linesWeek of 2026-04-26: +1,367 linesWeek of 2026-04-26: -592 linesWeek of 2026-05-03: +0 linesWeek of 2026-05-03: -0 linesWeek of 2026-05-10: +9 linesWeek of 2026-05-10: -9 linesWeek of 2026-05-17: +1,318 linesWeek of 2026-05-17: -401 linesWeek of 2026-05-24: +1,280 linesWeek of 2026-05-24: -36 linesWeek of 2026-05-31: +1,526 linesWeek of 2026-05-31: -1,057 linesWeek of 2026-06-07: +4,419 linesWeek of 2026-06-07: -1,755 linesWeek of 2026-06-14: +4,393 linesWeek of 2026-06-14: -580 linesWeek of 2026-06-21: +5,589 linesWeek of 2026-06-21: -685 linesWeek of 2026-06-28: +10,221 linesWeek of 2026-06-28: -3,026 linesWeek of 2026-07-05: +6,931 linesWeek of 2026-07-05: -1,007 linesWeek of 2026-07-12: +12,108 linesWeek of 2026-07-12: -6,222 linesWeek of 2026-07-19: +3,189 linesWeek of 2026-07-19: -1,246 linesWeek of 2026-07-26: +1,227 linesWeek of 2026-07-26: -340 linesWeek of 2026-08-02: +1,300 linesWeek of 2026-08-02: -230 linesWeek of 2026-08-09: +265 linesWeek of 2026-08-09: -185 linesWeek of 2026-08-16: +2,708 linesWeek of 2026-08-16: -272 linesWeek of 2026-08-23: +450 linesWeek of 2026-08-23: -88 linesWeek of 2026-08-30: +4,206 linesWeek of 2026-08-30: -939 linesWeek of 2026-09-06: +6,090 linesWeek of 2026-09-06: -4,625 linesWeek of 2026-09-13: +4,398 linesWeek of 2026-09-13: -3,469 linesWeek of 2026-09-20: +3,765 linesWeek of 2026-09-20: -363 linesWeek of 2026-09-27: +2,392 linesWeek of 2026-09-27: -431 linesWeek of 2026-10-04: +13,234 linesWeek of 2026-10-04: -838 linesWeek of 2026-10-11: +7,948 linesWeek of 2026-10-11: -1,733 linesWeek of 2026-10-18: +3,280 linesWeek of 2026-10-18: -807 linesWeek of 2026-10-25: +4,596 linesWeek of 2026-10-25: -1,085 linesWeek of 2026-11-01: +4,190 linesWeek of 2026-11-01: -577 linesWeek of 2026-11-08: +14,733 linesWeek of 2026-11-08: -882 linesWeek of 2026-11-15: +0 linesWeek of 2026-11-15: -0 linesNov 23, 2025Nov 15, 2026
+199.7K lines added, -85.4K removed over the last year.

Commits per week

last 52 weeks
720Week of 2025-11-23: 0 commitsWeek of 2025-11-30: 0 commitsWeek of 2025-12-07: 0 commitsWeek of 2025-12-14: 0 commitsWeek of 2025-12-21: 0 commitsWeek of 2025-12-28: 0 commitsWeek of 2026-01-04: 9 commitsWeek of 2026-01-11: 27 commitsWeek of 2026-01-18: 8 commitsWeek of 2026-01-25: 28 commitsWeek of 2026-02-01: 72 commitsWeek of 2026-02-08: 17 commitsWeek of 2026-02-15: 9 commitsWeek of 2026-02-22: 3 commitsWeek of 2026-03-01: 2 commitsWeek of 2026-03-08: 16 commitsWeek of 2026-03-15: 2 commitsWeek of 2026-03-22: 2 commitsWeek of 2026-03-29: 12 commitsWeek of 2026-04-05: 1 commitsWeek of 2026-04-12: 24 commitsWeek of 2026-04-19: 13 commitsWeek of 2026-04-26: 7 commitsWeek of 2026-05-03: 0 commitsWeek of 2026-05-10: 4 commitsWeek of 2026-05-17: 14 commitsWeek of 2026-05-24: 6 commitsWeek of 2026-05-31: 6 commitsWeek of 2026-06-07: 15 commitsWeek of 2026-06-14: 28 commitsWeek of 2026-06-21: 20 commitsWeek of 2026-06-28: 32 commitsWeek of 2026-07-05: 21 commitsWeek of 2026-07-12: 17 commitsWeek of 2026-07-19: 12 commitsWeek of 2026-07-26: 9 commitsWeek of 2026-08-02: 14 commitsWeek of 2026-08-09: 7 commitsWeek of 2026-08-16: 10 commitsWeek of 2026-08-23: 11 commitsWeek of 2026-08-30: 20 commitsWeek of 2026-09-06: 23 commitsWeek of 2026-09-13: 43 commitsWeek of 2026-09-20: 21 commitsWeek of 2026-09-27: 19 commitsWeek of 2026-10-04: 40 commitsWeek of 2026-10-11: 37 commitsWeek of 2026-10-18: 34 commitsWeek of 2026-10-25: 43 commitsWeek of 2026-11-01: 38 commitsWeek of 2026-11-08: 61 commitsWeek of 2026-11-15: 0 commitsNov 23, 2025Nov 15, 2026
857 commits in the last 52 weeks.

When work happens

weekday and hour
SunMonTueWedThuFriSat036912151821Sun 0:00 — 1 commitsSun 1:00 — 4 commitsSun 2:00 — 2 commitsSun 3:00 — 4 commitsSun 4:00 — 2 commitsSun 5:00 — 1 commitsSun 6:00 — 1 commitsSun 7:00 — 2 commitsSun 8:00 — 0 commitsSun 9:00 — 1 commitsSun 10:00 — 1 commitsSun 11:00 — 10 commitsSun 12:00 — 5 commitsSun 13:00 — 14 commitsSun 14:00 — 3 commitsSun 15:00 — 4 commitsSun 16:00 — 7 commitsSun 17:00 — 8 commitsSun 18:00 — 3 commitsSun 19:00 — 6 commitsSun 20:00 — 10 commitsSun 21:00 — 5 commitsSun 22:00 — 11 commitsSun 23:00 — 16 commitsMon 0:00 — 70 commitsMon 1:00 — 52 commitsMon 2:00 — 2 commitsMon 3:00 — 3 commitsMon 4:00 — 3 commitsMon 5:00 — 1 commitsMon 6:00 — 9 commitsMon 7:00 — 11 commitsMon 8:00 — 25 commitsMon 9:00 — 42 commitsMon 10:00 — 25 commitsMon 11:00 — 32 commitsMon 12:00 — 38 commitsMon 13:00 — 60 commitsMon 14:00 — 54 commitsMon 15:00 — 59 commitsMon 16:00 — 51 commitsMon 17:00 — 114 commitsMon 18:00 — 34 commitsMon 19:00 — 47 commitsMon 20:00 — 43 commitsMon 21:00 — 37 commitsMon 22:00 — 34 commitsMon 23:00 — 32 commitsTue 0:00 — 41 commitsTue 1:00 — 23 commitsTue 2:00 — 11 commitsTue 3:00 — 11 commitsTue 4:00 — 9 commitsTue 5:00 — 8 commitsTue 6:00 — 1 commitsTue 7:00 — 10 commitsTue 8:00 — 22 commitsTue 9:00 — 26 commitsTue 10:00 — 35 commitsTue 11:00 — 44 commitsTue 12:00 — 60 commitsTue 13:00 — 65 commitsTue 14:00 — 52 commitsTue 15:00 — 61 commitsTue 16:00 — 64 commitsTue 17:00 — 58 commitsTue 18:00 — 29 commitsTue 19:00 — 34 commitsTue 20:00 — 35 commitsTue 21:00 — 49 commitsTue 22:00 — 29 commitsTue 23:00 — 26 commitsWed 0:00 — 57 commitsWed 1:00 — 22 commitsWed 2:00 — 13 commitsWed 3:00 — 4 commitsWed 4:00 — 2 commitsWed 5:00 — 1 commitsWed 6:00 — 8 commitsWed 7:00 — 7 commitsWed 8:00 — 22 commitsWed 9:00 — 37 commitsWed 10:00 — 42 commitsWed 11:00 — 38 commitsWed 12:00 — 53 commitsWed 13:00 — 47 commitsWed 14:00 — 59 commitsWed 15:00 — 85 commitsWed 16:00 — 61 commitsWed 17:00 — 46 commitsWed 18:00 — 46 commitsWed 19:00 — 46 commitsWed 20:00 — 48 commitsWed 21:00 — 45 commitsWed 22:00 — 40 commitsWed 23:00 — 44 commitsThu 0:00 — 63 commitsThu 1:00 — 22 commitsThu 2:00 — 15 commitsThu 3:00 — 9 commitsThu 4:00 — 2 commitsThu 5:00 — 3 commitsThu 6:00 — 5 commitsThu 7:00 — 8 commitsThu 8:00 — 10 commitsThu 9:00 — 29 commitsThu 10:00 — 34 commitsThu 11:00 — 24 commitsThu 12:00 — 46 commitsThu 13:00 — 55 commitsThu 14:00 — 51 commitsThu 15:00 — 66 commitsThu 16:00 — 39 commitsThu 17:00 — 52 commitsThu 18:00 — 40 commitsThu 19:00 — 34 commitsThu 20:00 — 32 commitsThu 21:00 — 25 commitsThu 22:00 — 39 commitsThu 23:00 — 30 commitsFri 0:00 — 59 commitsFri 1:00 — 44 commitsFri 2:00 — 14 commitsFri 3:00 — 2 commitsFri 4:00 — 9 commitsFri 5:00 — 3 commitsFri 6:00 — 4 commitsFri 7:00 — 6 commitsFri 8:00 — 4 commitsFri 9:00 — 30 commitsFri 10:00 — 30 commitsFri 11:00 — 30 commitsFri 12:00 — 30 commitsFri 13:00 — 28 commitsFri 14:00 — 40 commitsFri 15:00 — 39 commitsFri 16:00 — 44 commitsFri 17:00 — 34 commitsFri 18:00 — 42 commitsFri 19:00 — 22 commitsFri 20:00 — 32 commitsFri 21:00 — 24 commitsFri 22:00 — 35 commitsFri 23:00 — 17 commitsSat 0:00 — 15 commitsSat 1:00 — 12 commitsSat 2:00 — 5 commitsSat 3:00 — 5 commitsSat 4:00 — 4 commitsSat 5:00 — 2 commitsSat 6:00 — 1 commitsSat 7:00 — 5 commitsSat 8:00 — 2 commitsSat 9:00 — 3 commitsSat 10:00 — 10 commitsSat 11:00 — 5 commitsSat 12:00 — 4 commitsSat 13:00 — 7 commitsSat 14:00 — 5 commitsSat 15:00 — 4 commitsSat 16:00 — 7 commitsSat 17:00 — 11 commitsSat 18:00 — 12 commitsSat 19:00 — 7 commitsSat 20:00 — 6 commitsSat 21:00 — 9 commitsSat 22:00 — 9 commitsSat 23:00 — 8 commits
Commit volume by weekday and hour (UTC). Larger dots mean more commits.
DateListRankStars gained
Sep 18, 2026daily#13+260
Sep 17, 2026daily#13+260
Sep 16, 2026daily#10+217
Aug 7, 2026daily#8+146
Aug 6, 2026daily#8+146
Jun 9, 2026daily#24+17
Jun 8, 2026daily#12+43
Jun 7, 2026daily#5+128