#155155th of 168 in ML frameworks and libraries
py-motmetrics
Python library for scoring multiple object tracking with MOTChallenge metrics
- Stars
- 1.5k
- License
- MIT
- Last commit
- Jul 2026
- Last release
- Oct 2024
- Language
- Python
1,475 on 21 Apr 20261,486 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
py-motmetrics evaluates multi-object tracker output against ground truth, computing CLEAR MOT (MOTA, MOTP), Identity (IDF1) and HOTA-related metrics. It reads MOTChallenge text, VATIC text and UA-DETRAC MAT/XML files, and returns a pandas-backed summary table. It can be called from Python or run as a command-line evaluator over a folder of sequences.
Who it is for: Computer vision researchers benchmarking multi-object trackers
Strengths
- Results aligned with the MOTChallenge devkit, with a parity test against TrackEval 1.3.0
- Auto-detects MOTChallenge, VATIC and UA-DETRAC input formats
- Summary keeps raw pandas data, exportable with to_csv
- MOTAccumulator and MetricsHost allow lower-level custom metric selection
Weaknesses
- MOTP reports average distance, so it needs conversion to match MOTChallenge percentages
- Evaluation only: it does not run or train trackers
- Last release 2024-10-30, well behind recent commits
What it needs
- no GPU
- Docker
- Needs pandas
Quick answers
Is py-motmetrics open source?
Yes. Its code is on GitHub under the MIT license.
Can you run py-motmetrics with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does py-motmetrics need a GPU?
No, it runs without one.
Also in ML frameworks and libraries
See all 168| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |