PAIR is currently based on three areas:

• Helping engineers and researches to build and understand machine learning systems better by applying design thinking and HCI practices to their work.
• Designing augmentative systems, so that professionals can work better with the help of AI.
• Making AI more transparent, inclusive and easier to understand for everyday users.

Google is planning to enable new research, education and open-sourcing new tools for designers and researches. Today they are releasing a software for researchers called Facets to analyse and visualise data sets before training AI models.

More details on the PAIR open source repository.

[caption id=”attachment_3733" align=”alignnone” width=”1440"]

Facets — displaying Quick, Draw! data[/caption]

Read more about it on their blog: blog.google

Image: Google

PAIR: Google’s Initiative to Bring Human-Centered Design and AI Research Together was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.

Giorgia Lupi on Twitter

I would keep scrolling this forever: https://t.co/QuNeausIjj "A casual platform for research design on generative drawings.

[caption id=”attachment_3656" align=”alignnone” width=”1280"]

Source: thedotisblack[/caption]

thedotisblack.tumblr.com

I Would Keep Scrolling This Forever was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.

inspirobot.me

It’s fun and addictive to keep clicking that ‘generate’ button and see what the bot is coming up with next. Twitter seems to love it too.

Images: InspiroBot

InspiroBot: Generating Hilarious Inspirational Garbage was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.

aiexperiments.withgoogle.com

More Resources
tensorflow.org
cloud.google.com/vision/
ml4a.github.io

A.I. Experiments: Making It Easier for Anyone to Explore A.I. was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.

The Art of Neural Networks — Mike Tyka was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.

Quick, Draw!

Quick, Draw! is a game built with machine learning. You draw, and a neural network tries to guess what you’re drawing. Let’s Play!

Earlier this year a team at Google’s Magenta project released Quick, Draw!, a fun little game where you can start drawing something and the app will try and guess what it is.

Since then, over 15 million players have contributed millions of drawings playing Quick, Draw! These doodles are a unique data set that can help developers train new neural networks, help researchers see patterns in how people around the world draw, and help artists create things we haven’t begun to think of.

Google is now open-sourcing this huge dataset for artists and developers to start training their own models.

quickdraw.withgoogle.com/data

[caption id=”attachment_3562" align=”alignnone” width=”900"]

Frame from https://youtu.be/2FAjQ6R_bf0[/caption]

Further Reading

Projects using the dataset

How do you draw a circle? We analyzed 100,000 drawings to show how culture shapes our instincts

Magenta is a Google Brain project to ask and answer the questions, “Can we use machine learning to create compelling art and music? Learn more

What Do 50 Million Drawings Look Like? was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.

The Pool and the Stream: A Podcast about Inspiration was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.

Colormind Blog
Color palette extraction

[caption id=”attachment_3439" align=”alignnone” width=”1024"]

Image: Colormind[/caption]

[caption id=”attachment_3448" align=”alignnone” width=”1280"]

Colormind webapp color browser[/caption]

colormind.io

A Deep Learning Color Scheme Generator was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.

Graphic Design’s Next Great Challenge: Branding AI was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.

Sketch-RNN is an assisted drawing experiment built on top of a neural network. Draw an object, and it will come up with many possible ways to continue drawing. Try it!

This experiment lets you draw together with a recurrent neural network model called Sketch-RNN. The developers taught this neural net to draw by training it on millions of doodles collected from the Quick, Draw! game, released earlier this year on Google’s AI experiments.

Once you start drawing an object, Sketch-RNN will come up with many possible ways to continue drawing this object based on where you left off. The model can also mimic your drawings and produce similar doodles.

We believe sketch-rnn will enable many creative applications. Even the decoder-only model trained on various classes can assist the creative process of an artist by suggesting many possible ways of finishing a sketch, helping artists expand their imagination. In the conditional model, exploring the latent space between different objects can potentially enable artists to find interesting intersections and relationships between different drawings. Even in the simplest use, pattern designers can apply sketch-rnn to generate a large number of similar, but unique designs for textile or wallpaper prints.David Ha, Douglas Eck

This is another cool example of how you can use machine learning in fun and creative ways and who knows, maybe this is the beginning of something big — AI assisting designers?

aiexperiments.withgoogle.com/sketch-rnn-demo

[caption id=”attachment_3378" align=”alignnone” width=”900"]

I start with a wheel, it finishes the ambulance[/caption]

Magenta is a Google Brain project to ask and answer the questions, “Can we use machine learning to create compelling art and music? Learn more

Project CreditsDavid Ha@hardmaruJonas Jongejan@halfdanjIan Johnson@enjalotDouglas Eck@douglas_eck

Assisted Drawing with a Neural Network was originally published in Inspiring AI on Medium, where people are continuing the conversation by highlighting and responding to this story.