Free online tools to generate, calculate,
convert, format, encode, and play.
 

Image Object Detector

Upload a photo and automatically detect the objects inside it. Each object is labelled and framed with a bounding box, powered by a DETR or YOLO model running locally via Transformers.js. Nothing is uploaded; your image never leaves your device.


Settings

Larger models are more accurate but slower to download and run.
Only objects detected with at least this confidence are shown.

Image

Loading model…
Privacy: The detection model is downloaded once from a public CDN and cached in your browser. All processing runs locally on your device; your image is never uploaded to any server.

Result


How It Works

This tool uses an object-detection transformer — either DETR (DEtection TRansformer) or YOLOS (You Only Look One-level Series) — executed directly in your browser with Transformers.js and ONNX Runtime (WebAssembly / WebGPU). The workflow is entirely client-side:

  1. Load model: The selected model is downloaded once from a public CDN and cached by your browser for future use.
  2. Detect: Your image is fed through the model, which predicts a set of objects, each with a class label, a confidence score and a bounding box.
  3. Draw: Boxes above your confidence threshold are drawn over the image on a canvas, colour-coded by label; you can download the annotated result.
Tips:
  • The first run downloads the model and can take a moment; later runs are fast thanks to caching.
  • Use YOLOS Tiny for a quick preview and DETR ResNet-101 for the best accuracy.
  • Lower the confidence threshold to surface more (but less certain) objects.
  • These models recognise the 80 common COCO categories (people, animals, vehicles, everyday items and more).

Embed This Util

You can embed this util on your own site as a widget. Adding ?embed=1 to the URL loads a compact version with just the tool itself; no header, menu, or documentation. Paste this snippet into your HTML:


    

Copy snippet Adjust the height to taste.



Feedback

Help us improve this page by providing feedback, and include your name/email if you want us to reach back. Thank you in advance.


Share with