# roryclear/clearcam

Add object detection, tracking, mobile notifications, and search to any security camera.

Repository: https://github.com/roryclear/clearcam
Canonical: https://ross.abutalabs.com/products/clearcam
Homepage: https://clearcam.org
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: object-detection, tinygrad, selfhosted, nvr, rtsp, yolov9, cctv, clip, ai
Last push: 2026-08-25T09:48:16+00:00

## Health v2 (maintenance only)
Score: 86/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 93, longevity 43
- inputs: {"age_days": 606, "days_push": 8, "days_rel": 51, "gap_med": 30, "n_releases_24m": 12}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1254, forks 89 (observed 2026-08-28T04:04:06.310010+00:00)

## What it is
Clearcam is a self-hosted Python NVR that adds AI object detection, tracking, mobile notifications, and semantic search to any RTSP security camera using tinygrad for local GPU inference. It includes a web UI and an optional iOS app, with premium cloud features for remote viewing and push notifications.

## Use cases
- detect trespassers on my security cameras with AI
- add object detection and tracking to my RTSP camera
- search recorded CCTV footage for a specific vehicle
- get phone notifications when a person is detected on camera
- self-hosted NVR alternative to Frigate with GPU inference
- count traffic passing my camera
- view my camera feeds remotely from my iPhone

## When to choose
- you have RTSP cameras and a GPU-capable Mac or Linux machine
- you want simple setup and local, private AI detection
- you want semantic search over camera events
- Frigate feels too complex or underuses your GPU

## When to avoid
- you need Windows support (ffmpeg fix still needed)
- you want CPU-only inference on cheap hardware
- you need mature integrations and a large community ecosystem
- you require remote mobile notifications without any cloud dependency

## Facets
- artifact type: application
- maturity: active
- function: computer-vision, object-storage, search-engine, http-server, machine-learning, video-processing
- domain: computer-vision, self-hosted, security, privacy, artificial-intelligence
- platform: python, self-hosted
- tags: nvr, rtsp, object-detection, object-tracking, yolov9, tinygrad, clip, cctv, mobile-notifications, video-search, macos, linux, ios

## Member repositories
- roryclear/clearcam (main) score 86

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:06.310010+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 2026-08-30T05:07:25.734814+00:00, confidence not recorded.
  - readme: https://github.com/roryclear/clearcam (fetched 2026-08-28T04:04:06.310010+00:00, sha 77d427167c08)
  - homepage: https://clearcam.org (fetched 2026-08-29T12:18:06.049010+00:00, sha 9518e9bc43af)
- Data as of 2026-08-30T08:39:29.467469+00:00.