BigGAN - AI Weirdness

Curse your valentine with these candy hearts

By Janelle Shane On February 08, 2021 - 4 min read

When I’ve generated candy [/this-time-i-didnt-train-a-neural-20-02-14?r=1lic4&utm_campaign=post&utm_medium=web&utm_source=copy] heart [/more-candy-hearts-by-neural-network-18-02-12?r=1lic4&utm_campaign=post&utm_medium=web&utm_source=copy] messages [/candy-heart-messages-written-by-a-18-02-09?r=1lic4&utm_campaign=post&utm_medium=web&utm_

Searching for Bernie

By Janelle Shane On January 26, 2021 - 4 min read

“Leaning Tower of Pizza”, BigGAN steered by CLIP using Big SleepI wrote earlier about DALL-E [https://www.aiweirdness.com/the-drawings-of-dall-e/], an image generating algorithm recently developed by OpenAI. One part of DALL-E [https://openai.com/blog/dall-e/]’s success is another algorithm called CLIP [https://openai.com/blog/clip/], which

When data is messy

By Janelle Shane On July 03, 2020 - 4 min read

There’s a story I tell in my book [https://youlooklikeathing.com] because it’s a great illustration of how AI gets the wrong idea about what problem we’re asking it to solve: Researchers at the University of Tuebingen trained a neural net to recognize images [https://medium.com/

A weirdly broken AI

By Janelle Shane On April 24, 2020 - 3 min read

The other day someone told me about a program [https://deepai.org/machine-learning-model/text2img] that will generate scenes to match a text description. I’m always excited to test out algorithms like this because the task of “draw anything a human asks for” is so hard that even state-of-the-art results

What is the opposite of guacamole?

By Janelle Shane On May 10, 2019 - 3 min read

Last week I attempted to illustrate some neural net-generated racehorses by turning to another neural net - this time, one that generates images, called BigGAN [https://aiweirdness.com/post/182322518157/welcome-to-latent-space]. Using Joel Simon’s ganbreeder.app [http://ganbreeder.app] interface, I’m able to see what BigGAN can generate

Welcome to Latent Space

By Janelle Shane On January 26, 2019 - 3 min read

I’ve written before about BigGAN [http://aiweirdness.com/post/178619746932/imaginary-worlds-dreamed-by-biggan], an image-generating neural net that Google trained recently. It generates its best images for each of the 1,000 different categories [https://gist.github.com/yrevar/942d3a0ac09ec9e5eb3a] in the standard ImageNet dataset, from goldfish to planetarium to toilet

The creepiest images generated by BigGAN

By Janelle Shane On October 31, 2018 - 3 min read

Writing to you from deep inside the Uncanny Valley, I present to you some hand-selected images generated by an algorithm called BigGAN [https://arxiv.org/abs/1809.11096]. This algorithm looks at example images of a bunch of different objects and then tries its best to figure out how to

Imaginary worlds dreamed by BigGAN

By Janelle Shane On September 30, 2018 - 4 min read

These are some of the most amazing generated images I’ve ever seen. Introducing BigGAN, a neural network that generates high-resolution, sometimes photorealistic, imitations of photos it’s seen. None of the images below are real - they’re all generated by BigGAN. Preprints of the BigGAN paper are here

Tag: BigGAN

Curse your valentine with these candy hearts

Searching for Bernie

When data is messy

A weirdly broken AI

What is the opposite of guacamole?

Welcome to Latent Space

The creepiest images generated by BigGAN

Imaginary worlds dreamed by BigGAN

Writing

Subscribe

Recent Posts

Botober 2024

Bonus: "Ignore all previous instructions" gets weirder

An exercise in frustration

Bonus: A unicorn goes downhill

Hidden 3D Pictures

Hidden sheep

Follow

Botober 2024

Bonus: "Ignore all previous instructions" gets weirder

An exercise in frustration