<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/my-doorbell-tells-me-who-s-at-the-door-before-i-look-and-the-recognition-runs-on-the-nas-in-my-cupb-gqyxrmlq7" -->

---
title: My doorbell tells me who's at the door before I look,...
description: A DIY smart doorbell setup runs facial recognition entirely on a home NAS, avoiding cloud subscriptions like Ring Protect or Nest Aware. The stack combines an...
canonical: https://daily.dev/posts/my-doorbell-tells-me-who-s-at-the-door-before-i-look-and-the-recognition-runs-on-the-nas-in-my-cupb-gqyxrmlq7
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: My doorbell tells me who's at the door before I look, and the recognition runs on the NAS in my cupboard | daily.dev
og:description: A DIY smart doorbell setup runs facial recognition entirely on a home NAS, avoiding cloud subscriptions like Ring Protect or Nest Aware. The stack combines an...
og:url: https://daily.dev/posts/my-doorbell-tells-me-who-s-at-the-door-before-i-look-and-the-recognition-runs-on-the-nas-in-my-cupb-gqyxrmlq7
og:image: https://api.daily.dev/og/posts/gQYxrMlq7.png
og:image:alt: My doorbell tells me who's at the door before I look, and the recognition runs on the NAS in my cupboard
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# My doorbell tells me who's at the door before I look, and the recognition runs on the NAS in my cupboard

**[XDA Developers](https://daily.dev/sources/xda-developers)** · 4 min read · 0 upvotes · 0 comments

## Summary

A DIY smart doorbell setup runs facial recognition entirely on a home NAS, avoiding cloud subscriptions like Ring Protect or Nest Aware. The stack combines an RTSP camera feed, Frigate for motion/person detection, DoubleTake as an orchestrator listening to MQTT events, and CompreFace or InsightFace for local facial recognition inference. Hardware acceleration via Intel QuickSync (11th/12th gen or N100 chips) keeps CPU load low, achieving 400-800ms end-to-end latency. The setup enables local automations like personalized welcome announcements and instant security snapshots via Home Assistant, all without sending video data to third parties.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.xda-developers.com/doorbell-tells-whos-at-door-before-look-recognition-runs-nas-in-cupboard>

## Questions this post answers

### How can I set up facial recognition for my doorbell camera without using a cloud subscription?

Run a local stack combining Frigate for motion detection via a low-res RTSP substream, DoubleTake as a middleman that listens to Frigate's MQTT events and pulls full-resolution snapshots, and CompreFace or InsightFace for facial embedding and matching against a local database. This runs on an ordinary NAS with no cloud upload, keeping video data private.

_daily.dev surfaces home lab guides for developers wiring up local-first automation stacks like this._

### What hardware do I need to run facial recognition AI on a NAS without it choking the CPU?

Hardware-accelerated video decoding is essential rather than optional, since software-only decoding can overload a typical NAS processor. Intel QuickSync, available on many iGPUs and Intel 11th, 12th, and N-series chips including the N100, handles decoding with near-zero CPU load, making it practical to also run CompreFace or InsightFace inference alongside continuous recording.

_Developers speccing home lab hardware can track chip and NAS compatibility notes on daily.dev._

### How fast is facial recognition notification compared to typical cloud doorbell alerts?

A local Frigate plus DoubleTake plus CompreFace/InsightFace pipeline delivers end-to-end latency of roughly 400 to 800 milliseconds from someone stepping onto the porch to a notification arriving, compared to cloud doorbells that can take around 10 seconds after a person has already walked away.

_Anyone comparing local versus cloud doorbell latency can find similar teardown details on daily.dev._

## Similar posts on daily.dev

- [I ditched my Ring doorbell for a $50 camera with local AI, and I'm saving $180 a year](https://daily.dev/posts/i-ditched-my-ring-doorbell-for-a-50-camera-with-local-ai-and-i-m-saving-180-a-year-14nmp8vs2) · XDA Developers · 0 upvotes · 0 comments
- [My security cameras now describe what they see, and not a single frame leaves the house](https://daily.dev/posts/my-security-cameras-now-describe-what-they-see-and-not-a-single-frame-leaves-the-house-ijsgs1xsl) · XDA Developers · 1 upvotes · 0 comments
- [Amazon’s Ring rolls out controversial, AI-powered facial recognition feature to video doorbells](https://daily.dev/posts/amazon-s-ring-rolls-out-controversial-ai-powered-facial-recognition-feature-to-video-doorbells-42dyzrs6q) · TechCrunch · 0 upvotes · 0 comments
- [I'd rather build a Raspberry Pi security camera than ever pay for Ring again](https://daily.dev/posts/i-d-rather-build-a-raspberry-pi-security-camera-than-ever-pay-for-ring-again-gwa2a09aw) · XDA Developers · 9 upvotes · 0 comments

---

Tags: [#docker](https://daily.dev/tags/docker), [#self-hosting](https://daily.dev/tags/self-hosting), [#facial-recognition](https://daily.dev/tags/facial-recognition)

[View this post on daily.dev](https://daily.dev/posts/my-doorbell-tells-me-who-s-at-the-door-before-i-look-and-the-recognition-runs-on-the-nas-in-my-cupb-gqyxrmlq7)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"My doorbell tells me who's at the door before I look, and the recognition runs on the NAS in my cupboard","url":"https://daily.dev/posts/my-doorbell-tells-me-who-s-at-the-door-before-i-look-and-the-recognition-runs-on-the-nas-in-my-cupb-gqyxrmlq7","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/my-doorbell-tells-me-who-s-at-the-door-before-i-look-and-the-recognition-runs-on-the-nas-in-my-cupb-gqyxrmlq7"},"datePublished":"2026-09-04T12:04:58.893Z","dateModified":"2026-09-13T19:22:46.312Z","description":"A DIY smart doorbell setup runs facial recognition entirely on a home NAS, avoiding cloud subscriptions like Ring Protect or Nest Aware. The stack combines an...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/ecb38d7d1669b4302af32ad02ebd1115?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/ecb38d7d1669b4302af32ad02ebd1115?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"XDA Developers","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"XDA Developers","logo":"https://media.daily.dev/image/upload/s--kCg6yyAP--/f_auto,q_auto/v1774964407/logos/xda-developers?_a=BAMAMiWQ0","url":"https://daily.dev/sources/xda-developers"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/my-doorbell-tells-me-who-s-at-the-door-before-i-look-and-the-recognition-runs-on-the-nas-in-my-cupb-gqyxrmlq7","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"docker,self-hosting,facial-recognition","timeRequired":"PT4M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"XDA Developers","item":"https://daily.dev/sources/xda-developers"},{"@type":"ListItem","position":3,"name":"My doorbell tells me who's at the door before I look, and the recognition runs on the NAS in my cupboard"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/my-doorbell-tells-me-who-s-at-the-door-before-i-look-and-the-recognition-runs-on-the-nas-in-my-cupb-gqyxrmlq7#faq","mainEntity":[{"@type":"Question","name":"How can I set up facial recognition for my doorbell camera without using a cloud subscription?","acceptedAnswer":{"@type":"Answer","text":"Run a local stack combining Frigate for motion detection via a low-res RTSP substream, DoubleTake as a middleman that listens to Frigate's MQTT events and pulls full-resolution snapshots, and CompreFace or InsightFace for facial embedding and matching against a local database. This runs on an ordinary NAS with no cloud upload, keeping video data private. daily.dev surfaces home lab guides for developers wiring up local-first automation stacks like this."}},{"@type":"Question","name":"What hardware do I need to run facial recognition AI on a NAS without it choking the CPU?","acceptedAnswer":{"@type":"Answer","text":"Hardware-accelerated video decoding is essential rather than optional, since software-only decoding can overload a typical NAS processor. Intel QuickSync, available on many iGPUs and Intel 11th, 12th, and N-series chips including the N100, handles decoding with near-zero CPU load, making it practical to also run CompreFace or InsightFace inference alongside continuous recording. Developers speccing home lab hardware can track chip and NAS compatibility notes on daily.dev."}},{"@type":"Question","name":"How fast is facial recognition notification compared to typical cloud doorbell alerts?","acceptedAnswer":{"@type":"Answer","text":"A local Frigate plus DoubleTake plus CompreFace/InsightFace pipeline delivers end-to-end latency of roughly 400 to 800 milliseconds from someone stepping onto the porch to a notification arriving, compared to cloud doorbells that can take around 10 seconds after a person has already walked away. Anyone comparing local versus cloud doorbell latency can find similar teardown details on daily.dev."}}]}
```

