Core Capabilities

A complete vision processing pipeline

From video ingestion, to AI processing, to structured feature output — every stage is independently configurable.

▶

Video Ingestion

RTSP and camera stream input with low-latency video decoding and multi-stream scheduling.

RTSP / ONVIF / GB28181 support
Hardware-accelerated decoding
Multi-stream management & timestamp alignment

AI Processing

Pluggable algorithms for detection, recognition, and tracking with parallel model inference.

Object / face / vehicle detection
Face recognition & feature embedding
Multi-object tracking
OCR / industrial defects / custom models

→

Feature Output

A unified vision feature structure delivered via REST API, webhooks, and event streams.

REST API
Webhook
Structured event streams

Algorithm Plugins

Algorithms are plugins, not the platform

Detectors, trackers, and recognizers all exist as plugins. Applications consume standardized features and never depend on a specific algorithm implementation.

Detector

face-det-v2
object-det

Tracker

multi-object
track-id

Recognizer

face-emb
custom

Vision Feature

A unified vision feature structure

The platform outputs a unified structured vision feature. Applications consume only standardized data, independent of algorithm implementation.

// Vision Feature — unified structured output
{
  "type": "face",
  "timestamp": "2026-07-23T10:42:18.412Z",
  "source": {
    "stream_id": "rtsp://camera-01/track1",
    "frame": 18472
  },
  "object": {
    "class": "person",
    "bbox": [320, 118, 488, 402],
    "embedding": "<512-dim feature vector>",
    "attributes": { "track_id": 17 }
  },
  "confidence": 0.964,
  "metadata": {
    "model": "face-det-v2",
    "device": "rk3588-edge"
  }
}

Get Started

See the platform in action

Explore the technical demo, or tell us about your use case to request a live walkthrough.

See the Demo Request Demo