Floor-Plan Auto-Detect (Vision) API
BetaAn AI vision model proposes a floor plan from a picture of it (PNG, JPEG or WebP): outline, centre, rooms, structures and main entrance in pixel coordinates, shaped as vastu-analyse input, with a confidence note.
Call it with POST /v1/ai/vastu-vision-detect. Input: specific input. Output: JSON, as a direct response. Each successful call costs metered, from 15 credits.
- Endpoint
- POST /v1/ai/vastu-vision-detect
- Input
- Specific input
- Output
- JSON
- Delivery
- Direct response
- Cost
- Metered, from 15 credits
- Compute class
- AI
Example response
Fields in data (inferred from the example)
plan· objectmain_entrance· nullnorth_hint· stringimage· objectconfidence· objecttier· string
Example response (illustrative)
Written by hand from the documented shape: AI readings are not captured automatically. Real text, token counts and credits differ.
Loading the example…
Request schema and every field: the AI interpretation reference.
Floor-Plan Auto-Detect (Vision) API: common questions
What does the Floor-Plan Auto-Detect (Vision) API return?
An AI vision model proposes a floor plan from a picture of it (PNG, JPEG or WebP): outline, centre, rooms, structures and main entrance in pixel coordinates, shaped as vastu-analyse input, with a confidence note. The response's data object includes plan, main_entrance, north_hint, image, confidence and tier. See the reference for an example response.
What input does Floor-Plan Auto-Detect (Vision) need?
Specific input.
How much does a Floor-Plan Auto-Detect (Vision) call cost?
Metered, from 15 credits per successful call; a failed call is not charged. Plans and top-ups are on the pricing page.
Can I try it before paying?
Yes. Every account gets sandbox keys and trial credits, so you can call it from your own code before choosing a plan.