Detect toilets

Toilet detection · model toilet-detector

POST/v1/drawings/detection/toilets

Detect toilets and plumbing fixtures in architectural floor-plan PDFs. Accepts a previously uploaded document_id, enqueues inference, and returns a job you poll for results. Detections are returned as bounding boxes in PDF coordinate space along with a per-detection confidence score.

Returns: A job to check202 Accepted
Usage: Element detectionper sheet
Results in: A list of toiletsresult.toilets
Auth: API keyX-API-Key header

Processing time (free tier)

Free-tier toilets detection typically completes in 2–4 minutes per job. Processing time scales mainly with page count and drawing complexity — a dense multi-page set takes longer than a single sheet.

Accuracy note: The toilet detector achieves a mAP@0.5 of 0.70 with an average recall of ~47%. It reliably detects standard upright fixture symbols but may miss rotated or non-standard representations. We recommend treating results as a starting point for review rather than a complete count.

Need it faster? Developer and Enterprise plans run detection on dedicated GPU infrastructure. Upgrade your plan

Example

Floor plan input — A sample floor plan before and after toilet detection: 22 toilets detected.

Floor plan with detected toilets — A sample floor plan before and after toilet detection: 22 toilets detected.

How it runs

Send a document_id
Get a job_id
Check your results or receive a webhook

Request

This endpoint requires a document_id. Upload your PDF first.

Body is JSON — this endpoint does not accept file uploads directly.

Field	Description
`document_id`string · UUIDRequired	ID of the uploaded PDF. Must belong to this account and not be expired.
`page_numbers`array of integersOptional	1-based page indices to scan. Omit to scan all pages. Out-of-range values are skipped by the worker but still counted.
`webhook_url`string · URLOptional	URL to POST the completed job payload to. Delivered on developer, pro, and enterprise tiers only.

Header	Description
`X-API-Key`Required	Your API key. A missing or invalid key returns 401.
`Content-Type`Required	`application/json`

UsageUsage is counted on submission: each page in page_numbers (or every page of the document when omitted) is one element detection, not only the pages that contain toilets. Send only valid page indices.

Responses

The job is enqueued immediately. Poll GET /v1/jobs/{job_id} until status is complete or failed.

Field	Description
`job_id`string · UUID	Use this to poll for results.
`status`string	Always queued on this response.
`poll_url`string	Path only — prepend `https://api.anchorgrid.ai` to build the full URL.

When status is complete and model is toilet-detector, the result field on the job contains:

Field	Description
`document_id`string	UUID of the source document.
`toilets`array	List of toilet detections. Each item has id, page, bbox, and confidence.
`toilets[].id`string	Stable identifier — "tlt_" + 12 hex chars.
`toilets[].page`integer	1-based PDF page index where the fixture was detected.
`toilets[].bbox`object	Axis-aligned bounding box in PDF coordinate space: x1, y1, x2, y2.
`toilets[].confidence`float	Model confidence score for this detection, between 0 and 1.
`toilets_found`integer	Count of items in the toilets array.
`pages_analyzed`integer	Number of pages the worker actually scanned.
`model_version`string	e.g. `toilet-detector-v1.0.0`
`processing_time_ms`integer	Wall time for the inference task.

Detections include a confidence score. Scores above 0.5 are high-confidence fixtures. Lower-confidence detections (0.12–0.5) may warrant manual review, particularly for rotated or non-standard fixture symbols.

Usage & rate limits

Each page scanned is one element detection (See pricing). Rate limit: tier RPM (5 / 60 / 120 / 300).

Plan	When the allowance is used up
`trial`	Trial allowance — `402 FREE_TIER_LIMIT_REACHED` when used up, `402 TRIAL_ENDED` once the trial is over.
`developer`	Monthly allowance, then overage up to your overage limit — `402 OVERAGE_LIMIT_REACHED` at the limit (`429 QUOTA_EXCEEDED` when the allowance is used up without overage billing).
`enterprise`	No allowance check.

Rate-limit 429s include retry_after_seconds in the body. Quota 429s and rate-limit 429s return the same status code — check the error body to distinguish them.

Errors

Status	What it means
401	Missing or invalid X-API-Key.
402	Trial allowance used up.
404	document_id not found or expired.
422	Validation error — malformed UUID or invalid body.
429	Rate limit or monthly quota exceeded.

Examples

curl -X POST https://api.anchorgrid.ai/v1/drawings/detection/toilets \
  -H "X-API-Key: <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{
    "document_id": "550e8400-e29b-41d4-a716-446655440000",
    "page_numbers": [1, 2, 3]
  }'

Response · 202

{
  "job_id": "7c9e6679-7425-40de-944b-e07fc1f90ae7",
  "status": "queued",
  "poll_url": "/v1/jobs/7c9e6679-7425-40de-944b-e07fc1f90ae7"
}

Completed job · result

{
  "document_id": "550e8400-e29b-41d4-a716-446655440000",
  "toilets": [
    {
      "id": "tlt_a1b2c3d4e5f6",
      "page": 1,
      "bbox": {
        "x1": 120.5,
        "y1": 340,
        "x2": 145.25,
        "y2": 380.75
      },
      "confidence": 0.86
    },
    {
      "id": "tlt_b2c3d4e5f6a1",
      "page": 1,
      "bbox": {
        "x1": 210,
        "y1": 180.5,
        "x2": 234.75,
        "y2": 220.25
      },
      "confidence": 0.72
    },
    {
      "id": "tlt_c3d4e5f6a1b2",
      "page": 2,
      "bbox": {
        "x1": 88,
        "y1": 412,
        "x2": 112.5,
        "y2": 452
      },
      "confidence": 0.62
    }
  ],
  "toilets_found": 3,
  "pages_analyzed": 2,
  "model_version": "toilet-detector-v1.0.0",
  "processing_time_ms": 11240
}

Cookies & analytics