Building a Real-Time Face Detection in HTML5

[eduardolundgren]

Computer Vision has the potential to yield non-invasive, accurate and low-cost solutions for tracking objects in a video. Video tracking is a critical component of most Augmented Reality applications, since the objects in both real and virtual worlds must be properly aligned with respect to each other in order to preserve the idea of the two worlds coexisting. Augmented reality and tracking applications for advertising and entertainment are gaining more space on the web environment.

The web browser environment is evolving fast, providing functionalities that can be explored and through them attractive modern solutions can be developed. Different devices such as mobile phones, notebooks, and even head-worn displays (Google Glass), provide an embedded web browser capable of running JavaScript and HTML5. The possibility to use this work as a cross-platform tracking library is a reality.

This article will demonstrate how to combine different HTML5 APIs: get user media, typed arrays, video and canvas, in order to create those applications.

[PaulKinlan]

This sounds really cool. I would love to see this.

At the end of this article I would love to see actionable advice for the developer so they could start planning on where to use this.

[eduardolundgren]

Definitely, adding actionable advices for the developers so they could start planning on where to use face detection in real life is a good suggestion.

Will share a draft as soon as I have something wrapped up.

@PaulKinlan, any suggestion for article deadlines?

[PaulKinlan]

@eduardolundgren it depends on your availability. As soon as possible is all I can say until you make an offer for when you are able to deliver it :)