norfair-enough 2.4.3

A maintained fork of tryolabs/norfair — lightweight Python library for real-time multi-object tracking.

The upstream Norfair repository is no longer actively maintained. This fork keeps the library alive for production use. There is no claim to be the official successor — just a pragmatic continuation. New maintainers are welcome.

Tracking players with moving camera	Tracking 3D objects

Installation

norfair-enough currently supports Python 3.10+.

For the minimal version, install as:

pip install norfair-enough

To install with optional dependencies:

pip install norfair-enough[video]    # Adds several video helper features running on OpenCV
pip install norfair-enough[metrics]  # Supports running MOT metrics evaluation
pip install norfair-enough[metrics,video]  # Everything included

Import note: The Python import remains unchanged — use from norfair import ... as before.

If the needed dependencies are already present in the system, installing the minimal version is enough for enabling the extra features. This is particularly useful for embedded devices, where installing compiled dependencies can be difficult, but they can sometimes come preinstalled with the system.

Features

• Any detector expressing its detections as a series of (x, y) coordinates can be used with Norfair Enough. This includes detectors performing tasks such as object or keypoint detection (see examples).

• Modular. It can easily be inserted into complex video processing pipelines to add tracking to existing projects. At the same time, it is possible to build a video inference loop from scratch using just the library and a detector.

• Supports moving camera, re-identification with appearance embeddings, and n-dimensional object tracking (see Advanced features).

• The library provides several predefined distance functions to compare tracked objects and detections. The distance functions can also be defined by the user, enabling the implementation of different tracking strategies.

• Fast. The only thing bounding inference speed will be the detection network feeding detections to the tracker.

Documentation

Getting started guide.

API reference.

Examples & demos

We provide several examples of how the library can be used to add tracking capabilities to different detectors, and also showcase more advanced features. All demos are available in the demos/ directory of this repository.

Some demos include Dockerfiles for reproducibility. If you have a GPU, install NVIDIA Container Toolkit for GPU passthrough. CPU-only usage is possible but may require dependency adjustments.

Adding tracking to different detectors

Most tracking demos are showcased with vehicles and pedestrians, but the detectors are generally trained with many more classes from the COCO dataset.

1. Ultralytics YOLO: object detection with YOLO11, tracking centroids or bounding boxes.
2. Ultralytics YOLO Pose: pose estimation with YOLO11-Pose, tracking human keypoints.

Advanced features

1. Re-identification (ReID) of tracked objects using appearance embeddings. This is a good starting point for scenarios with a lot of occlusion, in which the Kalman filter alone would struggle.
2. Accurately track objects even if the camera is moving, by estimating camera motion potentially accounting for pan, tilt, rotation, movement in any direction, and zoom.
3. Tracking of small objects, using SAHI: Slicing Aided Hyper Inference.

Benchmarking

1. Computation of MOT17 scores using motmetrics4norfair.

How it works

The tracker works by estimating the future position of each point based on its past positions. It then tries to match these estimated positions with newly detected points provided by the detector. For this matching to occur, the tracker can rely on any distance function. There are some predefined distances already integrated in the library, and the users can also define their own custom distances. Therefore, each object tracker can be made as simple or as complex as needed.

As an example we use Ultralytics YOLO to get the single point detections to use with this distance function. We just use the centroids of the bounding boxes it produces around cars as our detections, and get the following results.

On the left you can see the detection points, and on the right how the tracker tracks them assigning a unique identifier through time. Even a straightforward distance function like this one can work when the tracking needed is simple.

The library also provides several useful tools for creating a video inference loop. Here is what the full code for creating the previous example looks like, using Ultralytics YOLO:

import numpy as np
from ultralytics import YOLO

from norfair import Detection, Tracker, Video, draw_points

# Set up Ultralytics YOLO detector
model = YOLO("yolo11n.pt")

# Norfair Enough
video = Video(input_path="video.mp4")
tracker = Tracker(distance_function="euclidean", distance_threshold=20)

for frame in video:
    results = model(frame, verbose=False)[0]
    detections = [
        Detection(points=np.array([(x1 + x2) / 2, (y1 + y2) / 2]).reshape(1, 2))
        for x1, y1, x2, y2 in results.boxes.xyxy.cpu().numpy()
    ]
    tracked_objects = tracker.update(detections=detections)
    draw_points(frame, drawables=tracked_objects)
    video.write(frame)

The video and drawing tools use OpenCV frames, so they are compatible with most Python video code available online. The point tracking is based on SORT generalized to detections consisting of a dynamically changing number of points per detection.

Motivation

Modern object detectors are increasingly easy to use (e.g., Ultralytics YOLO), but adding robust multi-object tracking on top of them still requires stitching together detection, state estimation, and identity management. Norfair Enough was born to fill that gap: a modular tracking layer that works with any detector outputting (x, y) coordinates, and can be plugged into existing pipelines with minimal effort.

Comparison to other trackers

Norfair Enough's contribution to Python's object tracker library repertoire is its ability to work with any object detector by being able to work with a variable number of points per detection, and the ability for the user to heavily customize the tracker by creating their own distance function.

If you are looking for a tracker, here are some other projects worth noting:

• ByteTrack and BoT-SORT are high-performance MOT trackers that achieve strong results on MOT benchmarks. They are tightly coupled to specific detection architectures.
• Ultralytics built-in tracking provides integrated tracking (ByteTrack, BoT-SORT) when using YOLO models. Convenient if you only use YOLO, but not detector-agnostic.
• SORT and Deep SORT use Kalman filters like Norfair Enough, but are hardcoded to bounding-box tracking with a fixed distance function. Both are released under the GPL.
• OC-SORT improves on SORT with observation-centric momentum, handling occlusion better. Like SORT, it is box-only.
• supervision by Roboflow offers ByteTrack integration alongside annotation and dataset tools. Useful if you want a broader CV toolkit, but less customizable for tracking specifically.

Norfair Enough stands out by being detector-agnostic, supporting any point geometry (centroids, bounding boxes, keypoints, 3D points), and offering a BSD-3 license with no copyleft restrictions.

Benchmarks

These benchmarks were produced using the motmetrics4norfair demo script. Our CI runs MOT metrics regression tests on every pull request to prevent tracking quality regressions.

MOT17 and MOT20 results obtained using motmetrics4norfair demo script on the train split. We used detections obtained with ByteTrack's YOLOX object detection model.

MOT17 Train	IDF1 IDP IDR	Rcll	Prcn	MOTA MOTP
MOT17-02	61.3% 63.6% 59.0%	86.8%	93.5%	79.9% 14.8%
MOT17-04	93.3% 93.6% 93.0%	98.6%	99.3%	97.9% 07.9%
MOT17-05	77.8% 77.7% 77.8%	85.9%	85.8%	71.2% 14.7%
MOT17-09	65.0% 67.4% 62.9%	90.3%	96.8%	86.8% 12.2%
MOT17-10	70.2% 72.5% 68.1%	87.3%	93.0%	80.1% 18.7%
MOT17-11	80.2% 80.5% 80.0%	93.0%	93.6%	86.4% 11.3%
MOT17-13	79.0% 79.6% 78.4%	90.6%	92.0%	82.4% 16.6%
OVERALL	80.6% 81.8% 79.6%	92.9%	95.5%	88.1% 11.9%

MOT20 Train	IDF1 IDP IDR	Rcll	Prcn	MOTA MOTP
MOT20-01	85.9% 88.1% 83.8%	93.4%	98.2%	91.5% 12.6%
MOT20-02	72.8% 74.6% 71.0%	93.2%	97.9%	91.0% 12.7%
MOT20-03	93.0% 94.1% 92.0%	96.1%	98.3%	94.4% 13.7%
MOT20-05	87.9% 88.9% 87.0%	96.0%	98.1%	94.1% 13.0%
OVERALL	87.3% 88.4% 86.2%	95.6%	98.1%	93.7% 13.2%

Citing

For citations in academic publications, please reference the original Norfair project. Export your desired citation format (BibTeX or other) from the Zenodo entry.

License

Copyright © 2022, Tryolabs and © 2026, Akator GmbH and contributors. Released under the BSD 3-Clause.

This project is a fork of tryolabs/norfair.