# peiyunh/tiny

Tiny Face Detector, CVPR 2017

Repository: https://github.com/peiyunh/tiny
Canonical: https://ross.abutalabs.com/products/peiyunh-tiny
Homepage: https://cs.cmu.edu/~peiyunh/tiny
Language: MATLAB
License: NOASSERTION
License Family: other
Topics: face-detector, face-detection, scale-invariance, detection, detector
Last push: 2020-03-21T02:04:30+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": 3554, "days_push": 2357, "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 1143, forks 311 (observed 2026-08-28T04:03:44.943127+00:00)

## What it is
A MATLAB implementation of the Tiny Face Detector (CVPR 2017), a deep-learning face detector specialized in finding very small faces using scale, resolution, and context modeling. Includes training, testing, and demo code built on MatConvNet with pretrained ResNet/VGG models.

## Use cases
- detect tiny faces in images
- find faces in large group photos or selfies
- reproduce CVPR 2017 face detection results
- train a face detector on WIDER FACE
- benchmark face detection on small objects

## When to avoid
- you need a Python or production-ready detector
- you don't have MATLAB with GPU/CUDA setup
- you need actively maintained code or a permissive license
- you need real-time video face detection

## Facets
- artifact type: library
- maturity: maintenance
- function: computer-vision, machine-learning, image-processing
- domain: computer-vision, deep-learning, image-processing
- platform: windows
- tags: face-detection, tiny-faces, cvpr-2017, matconvnet, resnet, object-detection, research-code, matlab, gpu, linux, macos

## Member repositories
- peiyunh/tiny (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:44.943127+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 2026-08-30T06:34:56.968447+00:00, confidence not recorded.
  - readme: https://github.com/peiyunh/tiny (fetched 2026-08-28T04:03:44.943127+00:00, sha ef73caee5e61)
  - homepage: https://cs.cmu.edu/~peiyunh/tiny (fetched 2026-08-29T12:40:06.028352+00:00, sha f61a631a1c4e)
- Data as of 2026-08-30T08:39:29.467469+00:00.