trackers 2.2.0rc1

A unified library for object tracking featuring clean room re-implementations of leading multi-object tracking algorithms

trackers

trackers gives you clean, modular re-implementations of leading multi-object tracking algorithms released under the permissive Apache 2.0 license. You combine them with any detection model you already use.

https://github.com/user-attachments/assets/eef9b00a-cfe4-40f7-a495-954550e3ef1f

Install

You can install and use trackers in a Python>=3.10 environment. For detailed installation instructions, including installing from source and setting up a local development environment, check out our install page.

pip install trackers

install from source

By installing trackers from source, you can explore the most recent features and enhancements that have not yet been officially released. Please note that these updates are still in development and may not be as stable as the latest published release.

pip install https://github.com/roboflow/trackers/archive/refs/heads/develop.zip

Quickstart

Use the trackers CLI to quickly test how our tracking algorithms perform on your videos and streams. This feature is experimental; see the CLI documentation for details.

trackers track --source source.mp4 --output output.mp4 --model rfdetr-nano --tracker bytetrack

Tracking Algorithms

trackers gives you clean, modular re-implementations of leading multi-object tracking algorithms. The package currently supports SORT and ByteTrack. OC-SORT, BoT-SORT, and McByte support is coming soon. For comparisons, see the tracker comparison page.

Algorithm	MOT17 HOTA	MOT17 IDF1	MOT17 MOTA	SportsMOT HOTA	SoccerNet HOTA
SORT	58.4	69.9	67.2	70.9	81.6
ByteTrack	60.1	73.2	74.1	73.0	84.0
OC-SORT	—	—	—	—	—
BoT-SORT	—	—	—	—	—
McByte	—	—	—	—	—

Integration

With a modular design, trackers lets you combine object detectors from different libraries with the tracker of your choice.

import cv2
from rfdetr import RFDETRNano
from trackers import ByteTrackTracker

model = RFDETRNano()
tracker = ByteTrackTracker()

cap = cv2.VideoCapture("source.mp4")
while True:
    ret, frame = cap.read()
    if not ret:
        break

    frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
    detections = model.predict(frame_rgb)
    detections = tracker.update(detections)

run with Inference

import cv2
import supervision as sv
from inference import get_model
from trackers import ByteTrackTracker

model = get_model(model_id="rfdetr-nano")
tracker = ByteTrackTracker()

cap = cv2.VideoCapture("source.mp4")
while True:
    ret, frame = cap.read()
    if not ret:
        break

    result = model.infer(frame)[0]
    detections = sv.Detections.from_inference(result)
    detections = tracker.update(detections)

run with Ultralytics

import cv2
import supervision as sv
from ultralytics import YOLO
from trackers import ByteTrackTracker

model = YOLO("yolo11n.pt")
tracker = ByteTrackTracker()

cap = cv2.VideoCapture("source.mp4")
while True:
    ret, frame = cap.read()
    if not ret:
        break

    result = model(frame)[0]
    detections = sv.Detections.from_ultralytics(result)
    detections = tracker.update(detections)

License

The code is released under the Apache 2.0 license.

Contribution

We welcome all contributions—whether it’s reporting issues, suggesting features, or submitting pull requests. Please read our contributor guidelines to learn about our processes and best practices.