Repository files navigation

Detect objects in images right in your browser usingTensorflow.js! Currently takes ~800msto analyze each frame on Chrome MBP 13" mid-2014.

SupportsTiny YOLO, as of right now,tfjs does not havesupport to run any full YOLO models (and your user's computers probablycan't handle it either).

Check out the Live Demo

(You can only get so far with 1 FPS)

yarn add tfjs-yolo-tiny

npm install tfjs-yolo-tiny

importyolo,{downloadModel}from'tfjs-yolo-tiny';constmodel=awaitdownloadModel();constinputImage=webcam.capture();constboxes=awaityolo(inputImage,model);// Display detected boxesboxes.forEach(box=>{const{    top, left, bottom, right, classProb, className,}=box;drawRect(left,top,right-left,bottom-top,`${className}${classProb}`)});

If you're using a custom Tiny YOLO model or want to adjust the defaultfiltering cutoffs, you may do so by passing an additional optionsobject.

Example:yolo(inputImage, model, { classProbThreshold: 0.8 });

Returns an array of objects.

Returns aPromise that can resolve to atf.Model.

PR's are more than welcome! Perf improvement or better test coverageare probably the two biggest areas of immediate need. If you have thoughtson extensibility as well, feel free to open an issue!

yarn install

If you're running tests, make sure toyarn add @tensorflow/tfjs@0.7.0so that you you don't get tfjs package not found errors. If you're developing,make sure to remove tfjs as a dependency, as it'll start using thelocal version oftfjs intead of the peer version.

Note: Test coverage is poor, definitely don't rely on them to catch your errors.

yarn test

yarn build

Or during development, use watch mode, you can use thedemo appto test out changes.

yarn watch