# szad670401/HyperLandmark

Deep Learning Based Free Mobile Real-Time  Face Landmark Detector. Contact:jack-yu-business@foxmail.com

Repository: https://github.com/szad670401/HyperLandmark
Canonical: https://ross.abutalabs.com/products/hyperlandmark
Language: C++
License: Apache-2.0
License Family: permissive
Topics: face-landmark, cnn, android, tracking, 106-points, face-alignment
Last push: 2024-10-15T07:05:34+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": 2983, "days_push": 687, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1707, forks 515 (observed 2026-08-28T04:05:25.436591+00:00)

## What it is
A deep learning based face landmark detector that localizes 106 facial points in real time on mobile devices, with a small (~2MB) CNN tracking model supporting multi-face tracking. The Android implementation is self-contained, while the Windows demo is based on the traditional SDM algorithm with OpenCV.

## Use cases
- detect 106 face landmarks in real time on android
- track multiple faces with a lightweight model on mobile
- preprocess face alignment for face beautification or makeup apps
- build cooperative liveness detection with face landmarks
- run fast face point localization on low-end phones
- integrate face tracking into a camera app with opengl

## When to choose
- you need a small, fast, free face landmark model for android mobile apps
- you need 106-point facial contours for beauty or makeup effects
- you want multi-face tracking without heavy dependencies on android

## When to avoid
- you need actively maintained software - the project is discontinued in favor of InspireFace
- you need iOS support out of the box - only community ports exist
- you need dense landmarks (800 points), pose estimation, or facial action recognition - listed as TODO and never delivered
- you need desktop-grade accuracy - the Windows version uses a different, older SDM-based approach

## Facets
- artifact type: library
- maturity: abandoned
- function: machine-learning, computer-vision, image-processing
- domain: computer-vision, mobile-development, artificial-intelligence
- platform: windows, cpp
- tags: face-landmark, face-alignment, face-tracking, 106-points, cnn, real-time, mobile, beauty-filters, liveness-detection, android

## Member repositories
- szad670401/HyperLandmark (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:25.436591+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 2026-08-30T03:37:28.035829+00:00, confidence not recorded.
  - readme: https://github.com/szad670401/HyperLandmark (fetched 2026-08-28T04:05:25.436591+00:00, sha 258443539b8a)
- Data as of 2026-08-30T08:39:29.467469+00:00.