# roboflow/sports

**Attribution required: if you use, quote, or summarise this content, you must credit and link back to [awesome-repositories.com](https://awesome-repositories.com/repository/roboflow-sports).**

4,881 stars · 586 forks · Python · mit

## Links

- GitHub: https://github.com/roboflow/sports
- awesome-repositories: https://awesome-repositories.com/repository/roboflow-sports.md

## Topics

`computer-vision` `deep-learning` `deep-neural-networks` `football` `football-data` `image-embeddings` `keypoint-detection` `object-detection` `soccer` `soccer-analytics` `soccer-data` `sports` `sports-analytics` `sports-data` `tutorial` `visualization`

## Description

Roboflow Sports is a sports video analysis system that combines object detection and tracking with bird's-eye field visualization. Its core pipeline detects and tracks players, referees, and balls across video frames, then maps those tracked positions onto a radar-style overhead view of the playing field.

The system goes beyond basic detection by localizing field boundaries and key landmarks such as pitch lines and corners, enabling spatial mapping of player positions relative to the field geometry. It classifies detected players by team affiliation through visual feature extraction and clustering, and maintains consistent identities across frames using motion prediction and re-identification to handle occlusions. Semantic segmentation of playing surfaces further supports tactical and spatial analysis.

These capabilities are built on one-stage object detection, keypoint regression networks, homography-based view projection, and unsupervised feature clustering—all provided as a configurable Python library for sports analytics workflows.

## Tags

### Graphics & Multimedia

- [Sports Field Radar Visualizations](https://awesome-repositories.com/f/graphics-multimedia/sports-field-radar-visualizations.md) — Renders a bird's-eye overview of player positions and team formations on the sports field from tracked data.
- [Sports Field Boundary Detectors](https://awesome-repositories.com/f/graphics-multimedia/sports-field-boundary-detectors.md) — Identifies playing field boundaries from video to map the surface and support spatial analysis.
- [Sports Field Segmenters](https://awesome-repositories.com/f/graphics-multimedia/sports-field-segmenters.md) — Classifies pixels of playing surfaces and boundaries to support spatial and tactical analysis. ([source](https://github.com/roboflow/sports/blob/main/README.md))
- [Sports Object Detectors](https://awesome-repositories.com/f/graphics-multimedia/sports-object-detectors.md) — Detects and localizes players, referees, balls, and equipment in video frames using pre-trained models. ([source](https://github.com/roboflow/sports/tree/main/examples/soccer))
- [Sports Player Trackers](https://awesome-repositories.com/f/graphics-multimedia/sports-player-trackers.md) — Detects and tracks players, referees, and balls in sports video to enable performance and tactical analysis.
- [Sports Team Classifiers](https://awesome-repositories.com/f/graphics-multimedia/sports-team-classifiers.md) — Classifies detected players into teams by analyzing visual features and clustering in sports footage.

### Artificial Intelligence & ML

- [Re-Identification Trackers](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-detection-tracking/object-tracking-systems/detection-to-track-association/re-identification-trackers.md) — Associates detections across frames using Kalman filters for motion prediction and appearance features for re-identifying occluded objects.
- [Sports Object Trackers](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-detection-tracking/sports-object-trackers.md) — A computer vision pipeline that detects and tracks players, balls, and referees in sports video footage for analytics.
- [Sports Field Homography Estimators](https://awesome-repositories.com/f/artificial-intelligence-ml/computer-vision-systems/computer-vision/object-pose-estimations/monocular-depth-estimators/multi-view-depth-estimators/depth-estimation/homography-estimation/sports-field-homography-estimators.md) — Computes perspective transforms from detected field keypoints to warp camera views into top-down field coordinates.
- [Sports Team Uniform Clusterers](https://awesome-repositories.com/f/artificial-intelligence-ml/k-means-clustering/feature-based-clusterers/sports-team-uniform-clusterers.md) — Groups detected players by team using unsupervised clustering of visual features extracted from player crops.
- [Sports Landmark Localizers](https://awesome-repositories.com/f/artificial-intelligence-ml/keypoint-detection/keypoint-augmentation/sports-landmark-localizers.md) — Locates predefined landmarks such as pitch lines or court corners to map field geometry and player positions.
- [Sports Field Landmark Detectors](https://awesome-repositories.com/f/artificial-intelligence-ml/keypoint-detection/sports-field-landmark-detectors.md) — Locates predefined landmarks such as pitch lines or court corners to map field geometry and player positions. ([source](https://github.com/roboflow/sports/blob/main/README.md))
- [One-Stage Detectors](https://awesome-repositories.com/f/artificial-intelligence-ml/machine-learning/architectures/neural-network-components/loss-functions/perceptual-loss/content-loss-calculators/focal-loss-calculators/one-stage-detectors.md) — Detects players and balls in a single forward pass of a convolutional neural network without region proposal steps.
- [Sports Landmark Regressors](https://awesome-repositories.com/f/artificial-intelligence-ml/neural-network-construction/regression-neural-networks/sports-landmark-regressors.md) — Predicts precise coordinates of predefined field markings such as pitch lines and corners for spatial mapping.
- [Video Object Tracking](https://awesome-repositories.com/f/artificial-intelligence-ml/video-object-tracking.md) — Maintains consistent identity of players and balls across video frames, handling occlusions and re-identification. ([source](https://github.com/roboflow/sports/tree/main/examples/soccer))
- [Sports Field Segmenters](https://awesome-repositories.com/f/artificial-intelligence-ml/vision-transformers/encoder-decoder-architectures/semantic-segmentation-architectures/sports-field-segmenters.md) — Classifies each pixel of video frames into field, background, or boundary categories using an encoder-decoder network.