# cheind/py-motmetrics

:bar_chart: Benchmark multiple object trackers (MOT) in Python

Repository: https://github.com/cheind/py-motmetrics
Canonical: https://ross.abutalabs.com/products/py-motmetrics
Language: Python
License: MIT
License Family: permissive
Topics: object-tracking, object-detection, clear-mot-metrics, benchmark, metrics, tracker, mot, mot-challenge
Last push: 2026-07-17T11:48:07+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 93, release rhythm 8, longevity 100
- inputs: {"age_days": 3435, "days_push": 47, "days_rel": 672, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1487, forks 261 (observed 2026-08-28T04:04:52.009781+00:00)

## What it is
py-motmetrics is a Python library for evaluating multiple object tracking (MOT) results with MOTChallenge-aligned CLEAR MOT, Identity, and HOTA metrics such as MOTA, MOTP, and IDF1. It offers both a Python API and a command-line evaluator that can automatically detect common tracking file formats.

## Use cases
- evaluate multiple object tracker results against ground truth
- compute MOTA MOTP IDF1 metrics for MOTChallenge
- benchmark object tracking algorithms in Python
- compare predicted tracking files to ground truth annotations
- compute HOTA metrics for tracking
- evaluate tracker output from VATIC or UA-DETRAC formats

## When to choose
- you need standardized MOTChallenge-compatible tracking metrics
- you want a maintained Python library with both API and CLI evaluation
- you need to parse multiple tracking annotation formats automatically
- you want frame-level matching events for custom metric analysis

## When to avoid
- you need real-time tracking rather than offline evaluation
- you work with single-object tracking only
- you need metrics for other vision tasks like detection or segmentation

## Facets
- artifact type: library
- maturity: active
- function: benchmarking, computer-vision, data-science, analytics
- domain: computer-vision, machine-learning, data-science, testing
- platform: python, cli, cross-platform
- tags: multiple-object-tracking, motchallenge, clear-mot, hota, idf1, mota, tracker-evaluation, metrics

## Member repositories
- cheind/py-motmetrics (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:52.009781+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:33:47.677807+00:00, confidence not recorded.
  - readme: https://github.com/cheind/py-motmetrics (fetched 2026-08-28T04:04:52.009781+00:00, sha b221582f0af9)
- Data as of 2026-08-30T08:39:29.467469+00:00.