More R2-D2 than Rembrandt, the new resident artist of Carnegie Mellon University. Researchers at Carnegie Mellon University created an AI-powered robot that, according to a news release, can produce magnificent paintings on genuine canvas using only text inputs after ChatGPT stunned the world with its abilities.
What is FRIDA?
The AI-powered robotic arm with a paintbrush known as the FRIDA (Framework and Robotics Initiative for Developing Arts) robot can create original paintings using images, works of art, human input, and even music.
Similar to OpenAI’s ChatGPT and DALL-E 2, the robot uses AI models that produce text or graphics in response to input. By using its AI robotics technology to create real works of art, FRIDA goes one step further. The program has inspired CMU students and researchers. It was organized by Schaldenbrand in partnership with RI faculty members Jean Oh and Jim McCann.
The AI program FRIDA, which bears the name of the Mexican artist Frida Khalo, requires only the most fundamental inputs, such as text descriptions and already-existing images, but it is also capable of more complex operations. In one instance, the research team asked FRIDA to paint “Dancing Queen” by ABBA while playing the song.
The allowance for imprecision in FRIDA is one of its most remarkable features. FRIDA is allowed to make mistakes and change the rest of the painting accordingly. It uses an above camera to monitor its own progress, unlike most AI systems, which are built to be as accurate as possible. Speed is also unimportant because it takes hours to complete each painting.
After spending an hour learning how to use its paintbrush, FRIDA uses enormous vision-language models (such as OpenAI’s Contrastive Language-Image Pre-Training) created on enormous data sets that blend text and images downloaded from the internet to interpret the input. Some image-generation methods, like OpenAI’s DALL-E 2 method, use enormous vision-language models to create digital images.
FRIDA, a more intelligent bot than similar ones, analyzes its brushwork in real-time and makes adjustments as necessary. One of the most challenging technical challenges is to minimize the simulation-to-real gap or the discrepancy between what FRIDA composes digitally and what it paints on the actual canvas. To teach the simulator how to accurately depict the physical capabilities of the robot and the painting materials, FRIDA uses the real2sim2real idea.
Researchers that worked on the creation of FRIDA have called its artwork “whimsical and impressionistic.” A Ph.D. candidate at Carnegie Mellon University named Peter Schaldenbrand remarked of FRIDA, “FRIDA is a robotic painting system, but FRIDA is not an artist… FRIDA is not generating the ideas to communicate. FRIDA is a system that an artist could collaborate with. The artist can specify high-level goals for FRIDA, and then FRIDA can execute them.”
How FRIDA helps artists?
Nonetheless, the developers of this FRIDA believe that the project will allow artists and AI to work together rather than displace them. “We want to promote human creativity through FRIDA. For instance, I wanted to be an artist. Now, I can collaborate with FRIDA to express my ideas in painting.”
Even a color palette for a human to mix and furnish for the robot is provided by FRIDA on its computer screen. Automatic paint mixing is currently being developed by Jiaying Wei, a master’s student at the School of Architecture, and Eunsu Kang, a professor in the Machine Learning Department. “FRIDA is a robotics initiative to promote human creativity, rather than replacing it, by providing intuitive ways for humans to express their ideas using natural language or sample images.”
Further research will enhance FRIDA’s abilities and expand them to encompass sculpting. James McCann, a different faculty member, described FRIDA as a project that “investigates the interface of the human and artificial invention.” As McCann said, “FRIDA is using the kind of AI models that have been developed to do things like caption images and understand scene content and applying it to this artistic generative problem.”
Content Source: tomorrowsworldtoday.com
Cover Image Source: inceptivemind.com
IMPORTANT DISCLAIMER: All content provided on this website, any hyperlinked sites, social media accounts and other platforms is for general information only and has been procured from third party sources. We make no warranties of any kind regarding this content. None of the content should be interpreted as financial, legal, or other advice meant to be relied on for any purpose. Any use or reliance on this content is done at your own risk and discretion. It is your responsibility to conduct research, review, analyze, and verify the content before relying on it.