I was honoured to be chosen to give the last talk of the day for the Machine Learning track at the Google DevFest event in Singpore on 30-November-2019.

My talk, titled “Faces in the Browser using TensorFlow.js”, went through the various levels of code in TensorFlow.js; described the tasks involved in manipulating faces in the browser (using the attached webcam, or the phone’s own camera in the case of mobile); and also how these models are trained (broadly).

To avoid the talk getting too theoretical, I put together a demo that everyone in the room could try for themselves : I even had two helpers on stage to show how the embeddings (aka ‘QR codes’) for each of our faces could be clustered so that the machine could recognise the individuals later.

Because the demo is in Javascript, executed entirely in the user’s browser, you can examine the (trashy) code just by doing an ‘inspect’.

The slides for my talk (which include a link to the operating demo) are here :

Presentation Screenshot

If there are any questions about the presentation please ask below, or contact me using the details given on the slides themselves.

Presentation Content Example

PS: And if you liked the content, please ‘star’ my Deep Learning Workshop repo :: Star



Martin Andrews

{Finance, Software, AI} entrepreneur, living in Singapore with my family.



blog comments powered by Disqus