Request access

The engines are meant to sit inside the systems you already run, from a live camera feed to a folder on a studio workstation.

In

Your footage

  • RTSP / CCTV
  • Archives
  • Buckets

FaceAnon · Matting · Super-resolution

Out

Your systems

  • VMS
  • Data lake
  • DAM

None of these interfaces is available yet. They describe how we intend each engine to be used.

RTSP streaming

Planned
Input
Live camera and fleet video
Output
De-identified stream
Engine
FaceAnon

Batch REST API

Planned
Input
Archives and recorded footage
Output
Processed files and reports
Engine
FaceAnon

Python SDK

Planned
Input
Your data pipeline
Output
Frames, tracks, metadata
Engine
All engines

Cloud workers

Planned
Input
Image and video batches
Output
Mattes and enhanced frames
Engine
Applied Vision

Local CLI

Planned
Input
Studio files on disk
Output
Layers and uncertainty maps
Engine
Applied Vision

Sketches of the interfaces we are designing. Names and parameters will change.

# Planned live ingestion
aynvia stream \
  --engine faceanon \
  --in  rtsp://10.0.4.12/cam-07 \
  --out rtsp://0.0.0.0:8554/cam-07-anon

Start with the source.

Resolution, motion, lighting and compression all change what a vision engine can do with the footage.

Know the volume.

Hours of video per week, images per day, and how quickly results are needed.

Name the environment.

Cloud, on-premises or air-gapped. Where the footage is allowed to go decides the deployment.

Define a useful output.

The signals, formats and review steps your downstream systems rely on.

Tell us what you need to protect, extract or inspect. We'll tell you honestly whether our engines can help yet.

Request access