# aim-uofa/AdelaiDet

AdelaiDet is an open source toolbox for multiple instance-level detection and recognition tasks.

Repository: https://github.com/aim-uofa/AdelaiDet
Canonical: https://ross.abutalabs.com/products/adelaidet
Homepage: https://git.io/AdelaiDet
Language: Python
License: NOASSERTION
License Family: other
Topics: fcos, blendmask, abcnet, object-detection, instance-segmentation, ocr, text-recognition, meinst, text-detection, condinst, solo, boxinst, solov2, densecl, adelaidet
Last push: 2024-08-23T00:56:57+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2414, "days_push": 741, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3478, forks 654 (observed 2026-08-28T04:08:06.194903+00:00)

## What it is
AdelaiDet is an open-source Python toolbox built on Detectron2 that implements multiple instance-level detection and recognition algorithms such as FCOS, BlendMask, SOLOv2, CondInst, BoxInst, and ABCNet for text spotting. It ships pretrained model zoos, configs, training tools, and demos for object detection, instance segmentation, and scene text recognition.

## Use cases
- train object detection models like FCOS on COCO
- run instance segmentation with BlendMask or SOLOv2
- detect and recognize scene text in images with ABCNet
- train weakly supervised segmentation with BoxInst from box annotations
- deploy pretrained detection models with ONNX export
- benchmark anchor-free detection baselines

## When to choose
- you already use Detectron2 and want state-of-the-art instance-level detection algorithms
- you need research implementations of FCOS, BlendMask, SOLO, CondInst, or ABCNet with pretrained weights
- you need scene text detection and recognition in one toolbox

## When to avoid
- you need a lightweight production inference server without Detectron2 dependencies
- you want actively developed features - the project is largely in maintenance mode
- you work outside PyTorch/CUDA environments

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, computer-vision, image-processing, ocr
- domain: computer-vision, machine-learning, deep-learning, image-processing
- platform: python
- tags: object-detection, instance-segmentation, text-detection, text-recognition, detectron2, fcos, blendmask, abcnet, solo, boxinst, condinst, model-zoo, linux, gpu

## Member repositories
- aim-uofa/AdelaiDet (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:06.194903+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 2026-08-29T18:36:29.551200+00:00, confidence not recorded.
  - readme: https://github.com/aim-uofa/AdelaiDet (fetched 2026-08-28T04:08:06.194903+00:00, sha 0efa8e687a9b)
  - homepage: https://git.io/AdelaiDet (fetched 2026-08-29T09:30:31.243770+00:00, sha 1d4011e93a64)
- Data as of 2026-08-30T08:39:29.467469+00:00.