Tag: insta360

DSGVO-Pixeler — Privacy-First Video Anonymization for 4K Action Cam Footage

If you work with action-cam or cycling footage, you’ve probably faced the same dilemma: you want to share your 4K videos, but you can’t expose license plates or faces. DSGVO-Pixeler solves this locally, on your own machine, without sending anything to the cloud.

DSGVO-Pixeler is a Python tool built around YOLOv8 that automatically detects both license plates and faces and pixelates them with a true mosaic effect (not just blur). It’s designed for Apple Silicon and handles 4K video while preserving audio. The priority is privacy: the system is tuned to avoid missed detections, and you can add padding or larger pixel blocks to ensure sensitive details stay unreadable.

Key features:

  • Local processing only — nothing leaves your machine
  • 4K support with optional downscaled detection for speed
  • Apple Silicon acceleration with CPU fallback
  • Hardware encoding via ffmpeg VideoToolbox, or software fallback
  • Separate plate and face models, plus optional extra models
  • Configurable pixel strength, padding, and detection confidence
  • Optional no-pixel zones (HUD overlays) with pixel-accurate coordinates
  • Fast test runs (first N minutes) and visual debug overlays

Typical workflow:
1) Export your action-cam footage as a flat 16:9 video.
2) Place your models into models/plates/ and models/faces/.
3) Run DSGVO-Pixeler and get an anonymized MP4 with audio intact.

Example:

Python
python dsgvo-pixeler.py --input input.mp4 --weights models/plates/best.pt --preset balanced

DSGVO-Pixeler is practical for long rides and busy street scenes, where a single clip can contain hundreds of plates and faces. You stay in control: tune for accuracy with larger inference sizes, increase pixel strength for extra safety, or carve out no-pixel zones so your on-screen HUD stays crisp.

If you need a privacy-first pipeline for publishing public footage, DSGVO-Pixeler is a strong foundation that keeps everything local, fast, and configurable.

You find DSGVO-Pixeler on Github! – for free!

Examples:

Pixelate License Plates in 4K on Apple Silicon — Fast, Local, Privacy-First

EDIT: You will find a new version here: DSGVO-Pixeler

If you work with action-cam footage, sooner or later you face a tough challenge: how to blur license plates reliably without sending videos to the cloud. This project solves that problem locally, in 4K, on Apple Silicon.

Plater is a lightweight Python tool that detects license plates with YOLOv8 and pixelates them (true mosaic blocks, not blur). The pipeline is designed for privacy: it prefers false positives over missed plates and adds optional padding to every detection. That makes it practical for public road footage where privacy must come first.

What makes it useful:

  • 4K-capable detection with optional downscaled inference for speed
  • Apple Silicon acceleration (MPS) with CPU fallback
  • Hardware-encoded output via ffmpeg (VideoToolbox) with audio preserved
  • Configurable pixel strength, confidence threshold, padding, and no-pixel zones

Typical workflow:
1) Export your Insta360 footage as a reframed 16:9 flat video.
2) Run the CLI with your YOLOv8 plate model (.pt).
3) Receive an MP4 with plates pixelated and audio intact.

Example command:

Python
python blur_plates_m4.py --input input.mp4 --output output.mp4 --weights best.pt

If you work with long rides and hundreds of plates, the tool stays robust and flexible. You can trade speed for accuracy using work_w, or enforce a safety margin with pad. For overlays (speed/elevation HUD), you can define no-pixel zones so UI elements remain untouched.

Plater is fully local, scriptable, and ready for GitHub. If you need a privacy-first pipeline for public footage, this is a great starting point.