Middle of Nowhere වෙබ් අඩවි දෝෂ පිටුව

Press Kit

Unique Properties

This 404-page design comes from the idea of "The middle of nowhere." "404" means that the webpage doesn't exist, so this project creates an environment that makes users feel lost around the internet data and enjoy the virtual landscapes. By training a machine learning model that combines thousands of images, Yijia creates unique and beautiful places that don't exist in the whole world or even in the universe. They are just the landscapes generated by Artificial Intelligence.

Dimensions & Packaging

Responsive website construction

Tags

Machine Learning, Artificial intelligence, Website design, GAN

Production Technology

Generative adversarial network/Machine learning/UI Design

Design Challenge

It's difficult to collect all the appropriate databases and train the machine learning model. I tested many times and learned the new technology&knowledge during the model training to get unique and beautiful results.

Research

Our goal is to create an unprecedented 404 page connecting to IBM Research's brand identity and focus areas. I came up with different sketch directions, including typography&illustration style, and an interactive website. We finally decided on the concept of "Middle of nowhere" by using artificial intelligence technology. We hope to relay "Welcome to the future" to people through this 404 page.

Inspiration

As a top leader in the AI ​​industry, IBM Research is relentlessly curious about all the ways that computing can change the world. Today's AI is narrow. Applying trained models to new challenges requires an immense amount of new data training and time. People need AI that combines different forms of knowledge, unpacks causal relationships, and learns new things on its own. This 404 page explores a new possibility combining machine learning and design practice to increase the brand awareness of IBM research.

Project Duration

2021.11-2022.02 New York

Operation Flow

These images/videos of nonexistent landscapes were created using a generative adversarial network. It’s a type of machine learning model that can create novel content after being shown many original pieces of content. In this case, we showed our GAN thousands of images of landscapes from around the globe, and it generated these rather other-worldly images by combining aspects of everything it’s been shown.

Image Credits

IBM Research/Yijia Xie