# open-mmlab/mmdetection

OpenMMLab Detection Toolbox and Benchmark

Repository: https://github.com/open-mmlab/mmdetection
Canonical: https://ross.abutalabs.com/products/mmdetection
Homepage: https://mmdetection.readthedocs.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: object-detection, instance-segmentation, fast-rcnn, faster-rcnn, mask-rcnn, cascade-rcnn, ssd, retinanet, pytorch, panoptic-segmentation, rtmdet, semisupervised-learning, swin-transformer, transformer, vision-transformer, yolo, convnext, detr, glip, grounding-dino
Last push: 2024-08-21T02:01:07+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2933, "days_push": 743, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 32892, forks 9819 (observed 2026-08-28T04:11:57.912482+00:00)

## What it is
MMDetection is OpenMMLab's PyTorch-based toolbox and benchmark for object detection and instance/panoptic segmentation. It provides a large model zoo (Faster R-CNN, Mask R-CNN, YOLO, DETR, RTMDet, DINO and more), config-driven training, and inference tooling.

## Use cases
- train a custom object detection model on my own dataset
- run inference with pretrained Mask R-CNN or YOLO models
- benchmark detection models on COCO
- do instance segmentation in PyTorch
- fine-tune a pretrained detector for my images
- compare Faster R-CNN vs DETR implementations

## When to choose
- you need a mature, well-tested detection framework with a huge model zoo and reproducible benchmarks
- you want config-driven training/evaluation with modular customization of models, datasets, and pipelines
- you need state-of-the-art detectors (RTMDet, DINO, Co-DETR) with pretrained weights

## When to avoid
- you want a lightweight, minimal detector for edge deployment without heavy dependencies
- you prefer simple single-file training scripts over config-based frameworks
- you need non-PyTorch backends or very new research code not yet ported to MMDetection 3.x

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, computer-vision, image-processing, benchmarking
- domain: computer-vision, deep-learning, machine-learning, artificial-intelligence
- platform: python, cross-platform
- tags: object-detection, instance-segmentation, pytorch, model-zoo, mask-rcnn, yolo, detr, panoptic-segmentation, openmmlab, linux, gpu

## Member repositories
- open-mmlab/mmdetection (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:57.912482+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 2026-08-29T16:51:54.062871+00:00, confidence not recorded.
  - readme: https://github.com/open-mmlab/mmdetection (fetched 2026-08-28T04:11:57.912482+00:00, sha fc4002754ec4)
  - homepage: https://mmdetection.readthedocs.io (fetched 2026-08-29T07:48:49.069963+00:00, sha 874fa3809ecd)
- Data as of 2026-08-30T08:39:29.467469+00:00.