#coreml

Augmented Reality Style Transfer

Proof of concept experiment from Laan Labs lets you place a camera roll photo onto a wall and apply a neural net style transfer filter on it, including a convincing textured look:

“We’re taking the styles from well-known works of art, for instance van Gogh’s "The Starry Night”. The way it works is, we train a neural network to pickup stylistic components of existing paintings. The user’s photo is then transformed using the neural network so that it retains the overall structure but gains the stylistic elements of the image the network was trained on. Getting a realistic brush stroke texture was important here to be able to create the illusion of a real painting. So that took some effort to get right. 

You can see the original video at @madewitharkit here

Update 25/10/17

Laan Labs have uploaded these videos to YouTube:

Twit.AR

Coding experiment from Or Fleisher and Anastasis Germanidis combines Apple’s Augmented Reality and Machine Learning tools with Twitter to present connections which an iPad can ‘see’:

I chose to focus on spatializing information in Augmented Reality. The idea of spatializing information is not new in any sense and dates back to perhaps the invention of signage (or perhaps even earlier examples could be argued). With that being said, it seems advancements in accessibility of  Augmented Reality consumption models, predominantly the release of Apple’s ARKit and Google’s ARCore, calls upon the need of different approaches and models when spatializing information, or to be precise, drawing digital information in physical space ... Spatializing information in AR feels much more like cave glyphs than street signs, they are graphic, associative and story driven.

... in order to contextualize the real world to the digital world we need a bridge that allows us to understand some (very little but still) of the taught process that we go through between seeing things and thinking about ideas (yes, this ties perfectly into Peirce’s theory of signs and semiotics in general). To do that, we started by looking into another one of Apple’s new and  upcoming innovations CoreML. At it’s essence CoreML is an optimized engine for running machine learning (pretrained) models on iDevices. Apple also released quite a few pre-trained models themselves and so given our desire to classify objects from the real world we decided to use the Inception v3 model, which is trained to detect objects from images and classify them into a 1000 categories. 

AI‑Powered iOS App Development: Tools & Use Cases for 2025

Hello! In this article, you’ll learn about Core ML and Vision, two cutting-edge iOS frameworks. You’ll also get a quick overview of machine learning fundamentals.

https://vironit.com/machine-learning-in-ios/