Ir al contenido
  1. Todos los productos
  2. Todos los productos
  3. HR Attendance - Face Recognition Kiosk
-
HR Attendance - Face Recognition Kiosk

HR Attendance - Face Recognition Kiosk

https://armku.us/web/image/product.template/25/image_1920?unique=4ee3ead

$ 249.00 249.0 USD $ 249.00

Not Available For Sale

  • Odoo Version

Esta combinación no existe.

Odoo 16
Contáctanos

Terminos y Condiciones
Entrega instantanea por email - link de descarga de un solo uso
Perdiste el link? Te lo reenviamos gratis - responde a tu recibo
Agregar Premium Support para soporte prioritario y actualizaciones de version gratis

Compartir

At a glance

CategoryHuman Resources / Attendance
CompatibilityOdoo 16.0 · 17.0 · 18.0 Community & Enterprise — select your version above.
What ships~5 MB .zip per version — bundled face-api.js library + all 7 model weight files (tiny detector + 68-landmark + recognition). No manual download_models.sh, no SSH-side setup. Unzip, install, kiosk works.
Price$249.00 USD · one-time purchase, no recurring
Setup time~ 30–60 minutes for a typical Odoo install (unzip → restart → activate → enroll your first employee)
IncludedModule .zip with bundled model weights + Setup Guide PDF (attached to your receipt email) + email installation support for 7 days
Add-onsModule Installation & Setup (1h, $150) · Priority Support (1h within 90 days, $150)
LicenseLGPL-3 · install on any Odoo instance you operate
DeliverySingle-use signed download link emailed the moment your payment confirms. Link expires in 7 days; free re-issue if you lose it. See Terms.

What you get

Browser-based facial-recognition kiosk for Odoo HR Attendance. Employees clock in & out by standing in front of a camera-equipped tablet — no fobs, no badges, no cloud AI service, no PII leaving your network. Available natively for Odoo 16, 17, and 18.

How it works

Mount a tablet at the entrance, open the kiosk URL, and employees clock in by being recognized. face-api.js (TensorFlow.js underneath) runs the inference in the browser — no cloud API calls, no per-recognition cost, no photos leaving the device. Stored embeddings are 128 floats per employee, not photos, so they cannot be reversed into a face image.

Key features

  • Local inference — All recognition runs in the browser. No cloud AI service, no per-frame cost, no PII over the wire.
  • Per-employee enrollment — HR takes 1-3 photos from the employee form. Re-enroll anytime if appearance changes (glasses, haircut, lighting).
  • Configurable confidence threshold — default 70%, adjustable in Settings → Attendance → Face Recognition.
  • Health dashboard at /face_recognition/health — recent attendance events, cache status, force-refresh button.
  • Privacy by design — Embeddings are 128 floats per employee, not photos — irreversible. Optional retention policy on enrollment photos.
  • Standard hardware — Any USB webcam or built-in laptop camera. Any modern browser (Chrome 90+, Firefox 88+, Safari 14+) with WebRTC + WebGL.
  • Manager-only admin API — Cache status, force-refresh, recent face-recognition logs.

Architecture per version

  • Odoo 16 build — Legacy odoo.define JS modules (backend AbstractAction kiosk + FormController-based registration wizard). Battle-tested.
  • Odoo 17 build — Same architecture as v16, with the ensure_db import path updated to odoo.addons.web.controllers.utils. Legacy JS still loads via Odoo 17's @web/legacy compatibility shim.
  • Odoo 18 build (native) — Full rewrite. Vanilla ES6 IIFE modules served as plain <script> tags inside the rendered website templates. No odoo.define, no web.AbstractAction, no OWL component instantiation. Registration is a dedicated website page (/face_recognition/register/<employee_id>) opened from the employee form in a new tab. Health dashboard is a public-with-auth-check website page (/face_recognition/health). View XML migrated from the deprecated attrs="{...}" syntax to direct attribute form. Faster page load, fewer Odoo framework surprises on upgrades.

What's NOT included

  • The camera hardware (any USB webcam or built-in laptop camera works).
  • The tablet / kiosk hardware.
  • Cloud AI services — this module is local-inference only by design.
  • Liveness detection (anti-spoofing). A printed photo could trick the current matcher. Add a passive blink-check in v3 if your environment is high-risk.
  • Encryption at rest of the embeddings. They are stored as base64 JSON. Use Odoo's standard field-level encryption add-ons if your jurisdiction requires it.

Roadmap

  • Liveness detection — passive blink check (planned).
  • Multi-location kiosk routing — IP / geofence-based rules (planned).
  • OWL backend dashboard for v18 — replace the vanilla-JS health page with a richer in-backend OWL component (planned).