SightVision Pro
StableVision UnderstandingPARROTSIGHT nativeDeep visual understanding for enterprise video assets
P50 380msP95 720ms$1.20 / 1k calls
Overview
SightVision Pro analyses images and video frames at up to 4K resolution, producing structured scene descriptions, detected objects with bounding semantics, and natural-language summaries. It is trained on a curated ASEAN-scene corpus and optimised for CCTV, broadcast archive and industrial inspection workloads.
- Structured JSON scene graph output
- Natural-language captioning in EN / MS / ZH
- Batch frame analysis for video archives
- On-premise deployment available (Enterprise)
Capabilities
Scene accuracy96
Object recall93
ASEAN scene coverage91
Low-light robustness88
API
https://api.parrotsight.com/v1/vision/analyze
bash
curl -X POST https://api.parrotsight.com/v1/vision/analyze \
-H "Authorization: Bearer $PS_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"ps-vision-pro","image":"data:image/jpeg;base64,<BASE64_ENCODED_FRAME>","tasks":["scene","objects","caption"]}'Version history
- v2.4.1Latest2026-08-22
Improved low-light scene parsing; 9% latency reduction
- v2.3.0Released2026-06-30
Added Bahasa Malaysia captioning track
- v2.2.0Released2026-04-18
4K input support GA