# dlunion/DBFace

DBFace is a real-time, single-stage detector for face detection, with faster speed and higher accuracy

Repository: https://github.com/dlunion/DBFace
Canonical: https://ross.abutalabs.com/products/dbface
Language: Python
License Family: other
Last push: 2021-07-06T06:03:28+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2356, "days_push": 1884, "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 1355, forks 299 (observed 2026-08-28T04:04:29.178266+00:00)

## What it is
DBFace is a real-time, single-stage face detection model implemented in Python, offering small model sizes with high accuracy on the WiderFace benchmark. It supports export to ONNX and deployment via TensorRT and ncnn for fast inference.

## Use cases
- detect faces in images in real time
- run lightweight face detection on edge devices
- deploy a face detector with TensorRT or ncnn
- benchmark face detection accuracy on WiderFace
- train a custom small face detection model
- find small and hard-to-detect faces in crowded images

## When to choose
- you need fast, accurate face detection with a tiny model footprint
- you want a single-stage detector that outperforms RetinaFace-MobileNetV2 on hard faces
- you plan to deploy to mobile or embedded via ONNX/TensorRT/ncnn

## When to avoid
- you need general object detection beyond faces
- you require a maintained project with an active release cadence and a clear license
- you need facial landmark or recognition features out of the box

## Facets
- artifact type: library
- maturity: maintenance
- function: computer-vision, image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, deep-learning, artificial-intelligence
- platform: python, cpp, cross-platform
- tags: face-detection, object-detection, real-time-inference, onnx, tensorrt, ncnn, widerface, single-stage-detector, gpu

## Member repositories
- dlunion/DBFace (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:29.178266+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 2026-08-30T04:41:57.672906+00:00, confidence not recorded.
  - readme: https://github.com/dlunion/DBFace (fetched 2026-08-28T04:04:29.178266+00:00, sha 6919127473db)
- Data as of 2026-08-30T08:39:29.467469+00:00.