Proto 6/10: Copy printed text to desktop with AR+ML
Code: github.com/cyrildiagne/ar…
Book: Neurones, les intelligences simulées, Frédéric Migayroux & al (Editions Hyx 2018 @centrepompidou)
#ML #AR #AI #AIUX #ARCore #ARKit #WebXR
Technical Insights: ↓
Code: github.com/cyrildiagne/ar…
Book: Neurones, les intelligences simulées, Frédéric Migayroux & al (Editions Hyx 2018 @centrepompidou)
#ML #AR #AI #AIUX #ARCore #ARKit #WebXR
Technical Insights: ↓
The magic here is to use ARCore + AugmentedImages rather than SIFT.
Phone gets a new desktop screenshot on touch and adds it to ARCore (< 100ms).
Tracking is crazy fast & precise.
Interesting alternative to touch screen for interactive installations!
Phone gets a new desktop screenshot on touch and adds it to ARCore (< 100ms).
Tracking is crazy fast & precise.
Interesting alternative to touch screen for interactive installations!
Unexpectedly, it took me quite a couple iterations to figure out a good UI for the text selection flow.
The goal was to limit the thumb's movements while providing a precise bounding box to the Text Detection API.
In this case, the best iteration ended up being the simplest.
The goal was to limit the thumb's movements while providing a precise bounding box to the Text Detection API.
In this case, the best iteration ended up being the simplest.
Had to write this prototype in Java, and even though I didn’t particularly enjoy it, it was amazing to find everything I needed already available.
The more platforms you can “handle” (not master) the faster you can prototype, using the most optimal implementation.
The more platforms you can “handle” (not master) the faster you can prototype, using the most optimal implementation.
That’s it for this one! 📱📗🖥
Let me know if you have any question otherwise see you next week!
Let me know if you have any question otherwise see you next week!
