0.16.2

fixes:

• fixed issue of incorrectly initializing nodejs environment in electron renderer thread #157

other:

• bumped tfjs-core version to 0.14.2

0.16.1

fixes:

• detectSingleFace is now correctly returning face with highest score

0.16.0

• nodejs support

other:

• bumped tfjs-core version to 0.13.8

0.15.1

fixes:

• resolved issues with missing exports
• cleaned unused files from build folder

0.15.0

• new tiny face detection model for realtime face detection
• simplified and easier to use API + more utility (Composable Tasks API, FaceMatcher)

0.14.3

fixes:

• resolved broken dependency tree in package-lock.json, which caused tfjs-core to be bundled 3 times leading to ~3x bundle size + published fixed dist

0.14.2

• bumped tfjs-core version to 0.13.2, which comes with various performance improvements (speedup of ~2x for ssd mobilenetv1 face detector)

0.14.1

• published commonjs and es6 builds

0.14.0

• trained two 68 point face landmark detection models from scratch, which have higher accuracy and are much faster then previous model
• furthermore the model sizes are much smaller (350kb and 80kb tiny model) compared to the old model (7MB)

0.13.0

• major performance improvements by resizing net input canvases instead of tensors, which is much more performant and circumvents issue mentioned here

fixes:

• fixed post processing of 68 point face landmarks, which caused distortion of points at axes of minor dimension

breaking changes:

• removed managed flag and method from NetInput and related flag in toNetInput
• NetInput inputs are now left untouched, thus NetInput.inputs has been removed, NetInput.getInput(batchIdx) should be used instead
• NetInput and toNetInput do not accept tf.Tensor4D input with batchSize > 1 anymore, unstack batches instead and pass individual tensors as an array to create an equivalent batch input