In4Art’s Post

Thrilled to announce this Call For Proposals:  AI in Art: Practices, Processes, Doubts, and Revelations In-Person Roundtable at CAA 2025 in NYC Deadline for Proposals: August 29th Co-Chairs: Paul Catanese and Mat Rappaport This roundtable panel invites artists working with artificial intelligence(s) to reflect on their experiences, doubts, breakthroughs, and unresolved concerns, creating works using these technologies. We aim to hear from artists whose work and processes challenge popular narratives around "generative AI" tools like DALL-E, Midjourney, Stable Diffusion, and Adobe Firefly. We will consider the implications of "generative AI" on creativity through ethical, cultural, and societal lenses. Panelists will provide a brief overview of their projects, setting the stage for a dynamic, moderated discussion to foster new relationships, ideas, and community engagements. We encourage a broad spectrum of critical perspectives and aim to catalyze ongoing dialogues in and beyond the art community. We encourage proposals from artists with diverse practices - such as performance, installation, video, robotics, real-time environments, VR/AR, image, sound, text, and traditional materials- who are pioneering the use of AI in art. To submit proposals: https://lnkd.in/ga5zM7DT

Transforming Classics with AI: Bringing Art to Life The timeline is filled with AI animations, from classic ads to movie posters. Here's another exciting transformation! Transform the Classics and Bring Art to Life with AI Check out the famous painting by Leonardo da Vinci coming to life with the Chinese AI app, Kling. Why This Matters: -Innovative Use of AI: Leveraging AI to breathe new life into classic artworks. -Cultural Impact: Making art more accessible and engaging for everyone. -Technological Advancement: Showcasing the capabilities of AI in transforming traditional media. Stay tuned for more amazing AI innovations. Follow me, da sachin sharma, for the latest updates and insights in the world of AI. #AI #AIAnimation #ArtAndTech #LeonardoDaVinci #KlingApp #Innovation #TechInArt #DigitalTransformation

Researchers at The University of Texas at Austin have developed a generative AI model that transforms audio recordings into highly accurate street-view images, demonstrating the system’s ability to replicate the human connection between sound and visual perception. #ai #generativeai

Turning Sound into Sights: How AI Creates Street Views from Audio” 🌆 Imagine closing your eyes, listening to the world around you, and seeing vivid images come to life. Now, thanks to researchers at The University of Texas at Austin, AI can do just that! Using generative AI, the team has developed a model that turns audio recordings into accurate street-view images; blurring the line between human imagination and machine intelligence. Here’s how this groundbreaking innovation works: 🛠️ How They Did It: Researchers trained an AI model using 10-second audio clips and corresponding images from YouTube videos across urban and rural areas in North America, Asia, and Europe. The AI learned to “listen” to sounds; like traffic, voices, and leaves rustling—and generate high-resolution images from the audio alone. 🎯 The Results: The AI-generated images captured key details: 🔹 Proportions of sky, greenery, and buildings closely matched the real-world photos. 🔹 Architectural styles and object distances were surprisingly accurate. 🔹 Sounds like traffic or chirping insects helped the AI recognize whether it was daytime, cloudy, or nighttime. Human evaluators were 80% accurate in matching the generated visuals to the audio samples. 🧠 Why It Matters: This breakthrough proves that machines can replicate the deep sensory connections humans have with their environment. For example: 🌳 The rustling of leaves might signal a quiet forest. 🚗 The hum of traffic evokes a bustling cityscape. By combining sound and visuals, AI could help us better understand how we experience places—both physically and emotionally. 🛰️ Real-World Potential: Urban planning and noise mapping. Immersive virtual environments based on soundscapes. Enhancing accessibility for visually impaired individuals. 🔍 Looking Ahead: This innovation is just the beginning. As AI continues to evolve, it could redefine how we experience and recreate the world—one sound at a time. 📚 Source: UT Austin #AI #GenerativeAI #TechInnovation #SoundToImage #ArtificialIntelligence #Research #SmartCities #TechForGood

AI Daily Digest - Today's Strategic Insights Here are some of the AI headlines that caught our attention on June 19, 2024. ▶ Genuine Photo of Headless Flamingo Mistaken for AI Creation Wins Competition: A real photograph of a headless flamingo, titled “Flamingone,” won an AI-generated image competition but was later disqualified. Photographer Miles Astray entered the image to highlight the relevance of human creativity and the ethical implications of AI in photography. This incident is expected to spark discussions on AI use in photography. ▶ Snap Teases Real-Time, On-Device Image Diffusion Model for AR Experiences: At the Augmented World Expo, Snap introduced a real-time, on-device image diffusion model for augmented reality (AR). This model, small enough to run on a smartphone, will be integrated into Snapchat Lenses and available to AR creators by the end of the year. Snap’s new tools aim to significantly accelerate AR creation. ▶ Meta Unveils Five New AI Models and Research Initiatives: Meta has launched five major AI models, including multi-modal systems (Chameleon), next-gen language models with multi-token prediction, a text-to-music model (JASCO), AI speech detection (AudioSeal), and efforts to improve diversity in text-to-image models. These innovations aim to advance AI responsibly through open research and collaboration. #AIfuture #AIadvancement #AIandBusiness #wsiAI #AIforEveryone

Lighting isn’t just for studios and 3D anymore. AI is here to change the game. 💡✨ Imagine taking any portrait and changing its lighting as if it were shot under entirely new conditions—dramatic highlights, deep shadows, and dynamic effects, all without losing a single detail—and having full control over those lighting conditions. That’s the promise of SynthLight, a new AI-driven relighting method straight out of cutting-edge research. It’s powered by diffusion models and trained on synthetic 3D data to simulate lighting changes with incredible realism. While it’s still a research breakthrough, SynthLight shows us where AI is heading. Tools like this prove that what once seemed far off is now within reach—and it’s only going to get better from here. Lighting just became one more thing AI can master. Thoughts? Could you use this for your projects? 🔗 Project page and paper in the comments. #AI #Innovation #SynthLight #Relighting #FutureOfTech

Here is a new iteration in our current exploration revolving around the use of physical interfaces to feed and interact with a real-time img2img diffusion pipeline, using Stream Diffusion and SDXL Turbo. I may be spending way too much time messing around with this. 😊 #ai #stablediffusion #realtime #rnd #MR #mixedreality #aiart

Analog Meets AI: A New Creative Frontier [Part I] 🚨 The last few weeks have witnessed the intersection of genAI and analog tools, forging a new era of artistic expression. This paradigm shift allows creators to seamlessly blend physical and digital elements, giving birth to immersive and dynamic experiences. 🎨 In Ivan Yunitskiy latest experiment, 'AI ART + HUMAN, ', the controller is a human and its movements. "A motion capture sensor (Kinect) reads human poses and projects a generative shadow based on an AI model at a speed of 25 frames per second." 💡 Now imagine a concert where the artist's body becomes the primary sound and visual creation instrument. Their every gesture can be translated into data that feeds a generative AI model. This data is instantaneously transformed into visuals projected onto a massive screen behind the performer. A simple arm extension might trigger a cascading waterfall of colors, while a powerful leap could ignite a cosmic explosion. The result is a dynamic, always-changing, ever-evolving visual landscape that is entirely dictated by the artist's physicality. Link to source in the comments.

Text to 3D is here 🧊 This is Meta 3D Gen. A new system for generation of 3D assets from text in less than 1 min. "Meta 3D Gen is a new combined AI system that can generate high-quality 3D assets, with both high-resolution textures and material maps end-to-end, producing results that are superior to existing solutions — at 3-10x the speed of existing work in this space." AI keeps evolving. #genai #llms #ai #innovation #meta