Intro to Autolume for Generative AI

Autolume is a no-coding generative AI system allowing artists to train, craft, and explore their own models.

Joining us in Berlin by way of Vancouver, interdisciplinary artist and researcher Arshia Sobhan hosts a 2-day workshop (Feb 22 and 23, 10am-1pm) designed to introduce artists interested in training models with their unique datasets.*

*Requires Windows with an NVIDIA GPU (RTX 2070 or higher recommended

Autolume is a no-coding, user-friendly, visual synthesizer that leverages the artistic potential of GANs (Generative Adversarial Networks) without requiring technical skills. It streamlines the GAN workflow, including dataset preprocessing, model training, and real-time generation, making it accessible for non-technical users. Its unique features include interactive art generation through OSC messaging and the focus on a small-data approach to generative AI, enabling artists to use personal datasets for greater creative control. This approach not only avoids the biases of large pre-trained models but also offers an alternative to mainstream big data strategies, allowing for more personalized and unique artistic expressions away from conventional text-to-image practices.

Our workshop welcomes artists across all visual disciplines and technical backgrounds, whether you're keen on exploring generative AI for the first time or looking to deepen your engagement with more personalized and interactive generative AI workflows.

We particularly invite artists interested in training models with their unique datasets. Those participants will need to bring their images or videos (of their artworks or ethically sourced content). To ensure meaningful outcomes, we recommend bringing a minimum of 50-100 images or a ~5min video.

On the first day, participants will be guided through a comprehensive system demo and learn how to leverage Autolume for training models with their unique datasets. This session is crafted to empower artists to begin their journey into personalized generative AI creation, emphasizing the ease of use, level of control, and accessibility of Autolume for non-technical creatives.

The second day of the workshop shifts focus to the tangible outcomes of the previous day's endeavours, with a comprehensive review of the training results and a deep dive into Autolume's advanced features. Special attention will be given to its real-time and interactive capabilities, including audio-reactive and other forms of interactive art possibilities. This part of the workshop is designed to expand the creative possibilities for artists, offering insights into how Autolume can elevate their artistic endeavours to new heights.

What do I need?

We particularly invite artists interested in training models with their unique datasets. Participants will need to bring their images or videos (of their artworks or ethically sourced content).

To ensure meaningful outcomes, we recommend bringing a minimum of 50-100 images or a ~5min video.

Requires Windows with an NVIDIA GPU (RTX 2070 or higher recommended.

This event is brought to you by MANIFEST:IO, the symposium for new media and electronic art in Berlin. Check out event tickets and even secure a discounted ticket for the workshop using MAE.

Event tickets: https://maecommunity.app.link/RPCwlroLCPb

Host or Publisher TimeLab