Written Requirements:

Final Project presentations are demonstrations and do not require a slide deck, however, you might find slides useful to help you plan and structure your demo and discussion. Your documentation should come in written form (equivalent of 1-3 pages) as a blog post. It's up to you to figure out how to best document your project, at a minimum, please include the following:

Final Slides: https://www.canva.com/design/DAGYwE3xvm8/tqesUjLhmGjTjKyb75-KCg/view?utm_content=DAGYwE3xvm8&utm_campaign=designshare&utm_medium=link2&utm_source=uniquelinks&utlId=h7aba4a6f6e

Title: Surreal Visions

Sketch link: https://editor.p5js.org/TamaraBueno/full/4KRrnXdvR

One sentence description:

This project uses hand gestures and AI models to transform user inputs into dynamic, surreal visuals through an intuitive and interactive interface.

Project summary (250-500 words):

"Surreal Visions" is an interactive project that blends human creativity with AI-generated visuals, allowing users to create dynamic artwork through simple hand gestures or a digital canvas. The project uses machine learning models, such as Stable Diffusion, via the Replicate API, to generate unique, surreal images based on user inputs. By interacting through hand gestures, such as pinching the index and thumb together, users can select colors and submit their creations for AI interpretation. Users can also opt to use the camera for a more personalized experience or simply draw on the canvas.

Inspired by my experience visiting MoMA and seeing a generative art project by Refik Anadol, this project explores the intersection of AI and art, drawing from large-scale datasets similar to ImageNet. The goal is to create an immersive, fluid environment where visuals and gestures trigger captivating visuals, offering an intuitive, playful way to engage with AI-powered art creation. The user interface has been refined based on feedback, with improvements such as clearer instructions and the option to turn off the camera for privacy.

The experience is designed to be both fun and thought-provoking. It encourages creativity while providing a unique, interactive experience. Through user testing, it became clear that clear instructions and intuitive design were essential to making the experience accessible. If given more time, I would further enhance the project by integrating more interactive features, such as controlling the submission process with hand gestures or incorporating voice transcription for more immersive user interaction.

Overall, "Surreal Visions" is a fun, engaging, and creative exploration of AI’s potential in generating visual art, offering users a unique way to interact with technology while expressing their creativity.

Inspiration: How did you become interested in this idea? Quotes, photographs, products, projects, people, music, political events, social ills, etc:

• When one of my closest friends starting working for the MoMA, she took me a private tour of the MoMa and I could not forget this one piece. Having transitioned from computer science to Interactive Media Arts (IMA), I found this integration particularly fascinating. The project draws from 380,000 images of 180,000 artworks from MoMA’s collection, which reminded me of ImageNet’s large-scale dataset of over 14 million images and 21,841 synsets. While ImageNet is larger in scale, this project stands out for its unique connection to MoMA’s own art history, showcasing works that were previously exhibited at the museum.

Anadol’s Website & documentation on this project:

https://refikanadol.com/works/unsupervised/

Interesting Article I referenced about this Art piece:

https://amplify.nabshow.com/articles/machine-learning-moma-refik-anadol/

Refik Anadol. Unsupervised — Machine Hallucinations — MoMA. 2022.

Refik Anadol. Unsupervised — Machine Hallucinations — MoMA. 2022.

• In this class I wanted to explore more of using ML and visuals. My goal was to design a fluid, surreal environment that turns ambient sounds into captivating, reactive visuals. By receiving help in office hours I was able to brainstorm and come up with ideas on how to incorporate Stable Diffusion model via the Replicate API.

Process: How did you make this? What did you struggle with? What were you able to implement easily and what was difficult? [Down Below in Documentation]:

Audience/Context: Who is this for? How will people experience it? Is it interactive? Is it practical? Is it for fun? Is it emotional? Is it to provoke something?:

This project is designed for individuals interested in interactive generative art. Users can engage with it by pinching their index finger and thumb together to create drawings, selecting colors with the mouse, and deciding whether to use the camera or stick to the digital canvas when submitting an image to the model. The experience is meant to be fun and approachable, serving as a creative tool for those who might want an alternative to traditional software like Photoshop. By generating AI-powered visuals, the project aims to spark curiosity and provide a thought-provoking, personalized experience.

User testing: What was the result of user testing? How did you apply any feedback?:

I found many people were interest in drawing but they did not see the prompt button or the submit button. With no instructions it was kind of hard for them to know what they had to do. So my goal was to improve the UX/UI design of the project and provide instructions. I also notice some people did not want to be on camera, so I added a camera off button if they just wanted to use the canvas.

Source code [Down Below in Documentation]: