Annotate detections

POST /api/image-annotate-detections

Render bounding boxes and labels onto an image. Accepts the JSON output of any of the detection tools (faces / barcodes / QR / contours / etc.) and produces a visual overlay.

Draws coloured bounding boxes and text labels onto an image, given a list of detection rectangles. The companion tool for the various detection tools (`image-detect-qr`, `image-face-detect`, `image-detect-barcode`, etc.). Use the JSON output of any detection tool as the `detections` array — each entry needs `{rect: {x,y,width,height}, label?, color?}`.

Inputs

Name	Type	Default	Description
image*	file	—	Input image bytes.
detections*	string	—	JSON-encoded array of `{x, y, width, height, label?, score?, color?}` (or any object with `box: [x,y,w,h]`).
lineWidth	number (1…32)	3	Rectangle stroke width.
defaultColor	string	"#ff3366"	Stroke colour fallback (CSS hex).
fontScale	number (0…4)	0.6	Label font scale (0 = no labels).
format	enum (jpeg \| png)	"jpeg"	Output encoding.
quality	number (1…100)	90	JPEG quality.

Response

Modes: binary, json. Cache: yes (24h TTL).