contrast_renderer VS tfjs

This is true, but there is a lot more to the story. For one, WebGPU does not (yet) support mesh shaders, though it may later as an extension. For two, consider a glyph such as "o" that has two contours. Real triangulation generates a mesh that only generates triangles between the outer and inner contours, and mesh shaders aren't good at that. There are techniques (cover and stencil) that draw twice, incrementing and decrementing a winding number stored in the stencil buffer (see contrast renderer[1] for a clean modern implementation), but it does require nontrivial tracking on the CPU side, and can result in lots of draw calls to switch between the cover and stencil stages unless sophisticated batching is done.

Compute shaders avoid all these problems and work on WebGPU 1.0 today.

[1]: https://github.com/Lichtso/contrast_renderer

Thanks!

You can also checkout the "feature/ui" branch in GIT [1] to get the UI framework prototype.

[1]: https://github.com/Lichtso/contrast_renderer/tree/feature/ui

TensorFlow.js

Many people seem to be unaware of tensorflow.js, an official JS implementation of TF

https://github.com/tensorflow/tfjs

I'd love to see PyTorch in JS, but I think unless you get it running on the GPU it won't be able to do much.

TensorflowJS GitHub Repository

Also, Tensorflow.js WebGPU backend has been in the works for quite some time: https://github.com/tensorflow/tfjs/tree/master/tfjs-backend-...

It's a pity that tfjs never truly developed any decent ops. E.g. you need lgamma to implement the cap for zero-inflated poisson regression and tfjs simply doesn't have that: https://github.com/tensorflow/tfjs/issues/2011

People have been doing it for long with WebGL, see eg https://github.com/tensorflow/tfjs and https://cloudblogs.microsoft.com/opensource/2021/09/02/onnx-...

thanks, no not unity, going to show it as a demo with threejs + tensorflow on the web. I found a github request to add face orientation https://github.com/tensorflow/tfjs/issues/3835 looks like they assigned someone to add it but doesn't look like its available yet, but there's some posts about the math I can use to get rotations based on some of the landmarks

There already exist a wide variety of neural network inference engines that run in web browsers (e.g. TensorFlow.js and, my personal favorite for use with PyTorch models, ONNX Runtime Web), but pre- and post-processing has always required imperative manipulations on flat buffers rather than a clean ndarray interface.

Full docs and code: https://github.com/tensorflow/tfjs/tree/master/e2e/benchmarks/local-benchmark