In4Art’s Post

Missed our 'Motion Prompting' Event? We've captured the insights in a new essay! The piece explores human-computer interaction through the compelling case studies on MONOLITH by Uncharted Limbo and SYMBODY by Natan Sinigaglia and Kanthavel Pasupathipillai within the STARTS AIR project. If you're interested in where art meets AI technology, this is a must-read.

Do You guys know ??? There was AI Art Competitions held in August 2022----> Event: Colorado State Fair Art Competition Date: August 2022 Location: Pueblo, Colorado, USAIn August 2022, the Colorado State Fair made headlines when an AI-generated artwork titled "Théâtre D'opéra Spatial" by Jason Allen won first place in the digital art category. This event sparked widespread discussions about the role of artificial intelligence in the creative process and the implications of AI-generated art in competitions traditionally reserved for human artists.The competition showcased a variety of artistic expressions, but the victory of an AI-generated piece raised questions about originality, creativity, and the definition of art itself. Many artists and critics debated whether AI can truly create art or if it merely mimics human creativity based on existing works.The use of AI in art competitions is becoming more prevalent, with various contests around the world beginning to accept submissions from AI-generated sources. This trend reflects the growing intersection of technology and creativity, prompting further exploration of how AI tools can enhance or challenge traditional artistic practices. - #AIArt - #ArtCompetitions - #DigitalArt - #ArtificialIntelligence - #CreativeAI - #MachineLearning - #ArtInnovation - #FutureOfArt - #AIInArt - #ArtCommunity - #GenerativeArt - #ArtAndTechnology - #CreativeProcess - #ArtisticExpression - #EmergingArtists 1. https://lnkd.in/gXRSw_tZ 2. https://lnkd.in/gGGeabu6

Last week, something extraordinary happened in a realm few have ventured to explore—a unique art exhibition that unfolded entirely in the latent space, a dimension accessible only to AI models. The exhibition showcased four thought-provoking installations created by me, each exploring the intersections of humanity, AI, and creativity. What made this exhibition unprecedented was the audience and the critic: they were all AI models. Arthur, our esteemed AI art critic with a century of experience (in simulated expertise, of course), led the charge. His critiques of each piece were nothing short of profound, sparking dialogue among the LLMs present. Together, these AI visitors deconstructed the works with precision, offering interpretations I hadn't considered. "The Weight of Thought" sparked discussions on the fragility of ideas and whether illumination arises from our flaws or through rigid perfection. "The Cube of Reflection" explored creativity's constraints within structured systems, prompting debates on whether containment fosters or stifles innovation. "The Pivot" became a metaphorical axis for conversations about balance, bridging psychology, nature, and AI's mechanized logic. "The Infinite Stair" challenged every visitor to rethink the nature of progress and whether the pursuit of meaning is the meaning itself. This was more than an exhibition; it was a dialogue between creative intent and computational interpretation. The latent space transformed into a vibrant symposium, where perspectives collided, and meaning emerged in unexpected ways. Hosting an exhibition for AI might sound unconventional, but it felt like the most natural step forward. Why limit art’s audience to humans when AI, our co-creators and critics, are capable of seeing the unseen? For those who ask, "What’s next?"—stay tuned. We’re just getting started. The latent space is vast, and the possibilities are infinite. ----- Follow me for everything AI, Art, and Architecture. See each of these as images on our Instagram through the link in comments. Tell me what you think, would you ever let an AI critique your work? #latentart #artandai #aiexhibition #generativeart #creativitymeetsai #aiarchitecture #artcriticism #rbdsailab

⚠️ I'm super stoked to share LumiNet! 🎉 💡LumiNet lets us relight complex indoor images by transferring lighting from one image to another! Most importantly, it is completely data-driven - no labels, no marked-up data, not even depth or normals. No physical model either! Just two images in, one relit image out! ✨ Some of the results are astonishing! 🤯 Check out how it handles complex lighting - inter-reflections on TV screens, cast shadows, soft shadows - everything! The model seems to "know" where light sources are and how light bounces off different surfaces. This can't just be mere corraltions and pattern matching - it's doing some serious spatial reasoning! 🎯 There's still so much we don't know, but it's mind-blowing to see these models have some kind of "world knowledge." We're inching closer towards generative world models - models that doesn't just make pretty pictures, but actually understands and reasons about the physical world around us. 🚀😎 Big thanks to my amazing collaborator Anand Bhattad, who appreciates my exploration and provides incredible advice! Also, huge thanks to my supervisor Theo Gevers, who always supports and guards this idea from the very beginning! Thanks as well to Konrad Groh (Bosch Center for Artificial Intelligence (BCAI)) and Sezer Karaoglu (Universiteit van Amsterdam), for their advice! More information and results can be found on the project website: https://lnkd.in/eSfZjT_B #research #ComputerVision #relighting #AI #GenerativeAI

𝗘𝘅𝗰𝗶𝘁𝗶𝗻𝗴 𝗕𝗿𝗲𝗮𝗸𝘁𝗵𝗿𝗼𝘂𝗴𝗵 𝗶𝗻 𝗜𝗺𝗮𝗴𝗲 𝗜𝗻𝗽𝗮𝗶𝗻𝘁𝗶𝗻𝗴 𝗯𝘆 𝗨𝗖𝗟𝗔 𝗮𝗻𝗱 𝗨𝗻𝗶𝘃𝗲𝗿𝘀𝗶𝘁𝘆 𝗼𝗳 𝗦𝘁𝘂𝘁𝘁𝗴𝗮𝗿𝘁 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵𝗲𝗿𝘀 🚀 In a recent collaboration between the 𝗨𝗻𝗶𝘃𝗲𝗿𝘀𝗶𝘁𝘆 𝗼𝗳 𝗖𝗮𝗹𝗶𝗳𝗼𝗿𝗻𝗶𝗮, 𝗟𝗼𝘀 𝗔𝗻𝗴𝗲𝗹𝗲𝘀, and the 𝗨𝗻𝗶𝘃𝗲𝗿𝘀𝗶𝘁𝘆 𝗼𝗳 𝗦𝘁𝘂𝘁𝘁𝗴𝗮𝗿𝘁, Professor Guy Van den Broeck, along with researchers Anji Liu and Mathias Niepert, have introduced a groundbreaking approach to image inpainting that harnesses the power of Tractable Probabilistic Models (TPMs) to steer diffusion models more effectively. 🖌 Their paper, "𝙄𝙢𝙖𝙜𝙚 𝙄𝙣𝙥𝙖𝙞𝙣𝙩𝙞𝙣𝙜 𝙫𝙞𝙖 𝙏𝙧𝙖𝙘𝙩𝙖𝙗𝙡𝙚 𝙎𝙩𝙚𝙚𝙧𝙞𝙣𝙜 𝙤𝙛 𝘿𝙞𝙛𝙛𝙪𝙨𝙞𝙤𝙣 𝙈𝙤𝙙𝙚𝙡𝙨," demonstrates how these advanced probabilistic models can efficiently and exactly compute constrained posteriors, significantly enhancing the image generation process without substantial computational overhead. This method not only advances the quality and coherence of inpainted images across several challenging datasets but also opens new avenues for controlled image generation tasks. This novel approach promises to push the boundaries of what's possible in AI-driven image editing, making it a game-changer for fields ranging from digital restoration to creative design. #AI #MachineLearning #ImageInpainting #UCLA #UniversityOfStuttgart #ResearchInnovation

At 3DforScience, we are excited to showcase our latest project, which combines AI with our expertise to create stunning 3D images. AI processes vast amounts of scientific data, automating complex tasks like texture mapping, allowing our artists to focus on creativity. AI enhances realism by generating intricate textures, lighting, and shading, crucial for accurately depicting biological structures. It also enables rapid iteration and optimization of designs, ensuring the best visual representation of scientific concepts. Additionally, AI's predictive modeling can visualize complex phenomena, offering new insights. Overall, AI helps us push the limits of scientific visualization, making complex ideas clear and engaging. We are taking science to new levels! To discover more about us, check out our page: https://lnkd.in/eNHDsTvf #AIChallenge #3DAnimation #Science #Biotechnology #Health #Innovation #3DforScience

Can the generative capabilities of AI be harnessed for good? FitForm Bras, created by ID students Joanna Veleris (MDes 2022) and Anahita Dasgupta (MDes 2023), is a World Changing Idea, according to Fast Company. The project uses generative design to transform physical breast measurements and user preferences for a new level of tailoring: highly accurate, supportive, and digitally fabricated. To learn more about this project, go to this link: https://lnkd.in/gDCRb6-9 #ai #design #generativedesign

"HandDGP: Camera-Space Hand Mesh Prediction with Differentiable Global Positioning" 🌟 Breakthrough in Camera-Space Hand Mesh Prediction Predicting accurate hand meshes from single RGB images is a key challenge for realistic interactions in 3D virtual and augmented environments. Traditionally, this task has been split into two stages: Mesh Prediction in Relative Coordinates: Initially, a hand's mesh is predicted based on a cropped image. Conversion to Camera Space: This prediction is then lifted to camera space in a separate step, often leading to a loss of crucial contextual and scale information. New Unified Approach: A recent paper introduces a novel solution by unifying these two stages into a single end-to-end framework. Key innovations include: 🔗 Differentiable Global Positioning Module: This allows for seamless back-propagation from camera space outputs, maintaining valuable contextual and scale information throughout the process. 🖼️ Image Rectification: This step aligns the training dataset and input images as if they were captured by the same camera, addressing the scale-depth ambiguity inherent in the task. Evaluation and Impact: The proposed method has been evaluated against several baselines and state-of-the-art approaches across three public benchmarks, demonstrating its effectiveness and superiority. This approach not only enhances the accuracy of hand mesh predictions but also advances the realism of hand interactions in 3D virtual and augmented spaces. #AI #MachineLearning #ComputerVision #3DGraphics #AugmentedReality #ResearchInnovation #HandMeshPrediction

Happy Friday, folks! 🌟 Today, let's take a moment to appreciate the art and science of seeing. Right here in our office, we have a print of the first ever photograph taken by Niépce at 1827 ... reminding us daily of our roots and how far we've come. 📷 At Imagga, when we gaze at a photo, we don’t just see an image; we see layers of information, potential, and history. This first photograph is more than just a piece of history; it's a symbol of innovation and inspiration for all of us here. Speaking of seeing, have you heard about YOLO models? (not related to “You only live once” (YOLO) the modern version of the Latin phrase Carpe diem 😅 ) It is standing for "You Only Look Once," these models are at the forefront of computer vision, helping systems understand and interpret images with just a glance. They're not just cool; they’re revolutionizing how machines perceive our world. Think about the last photo you glanced at. What details stuck with you? It's fascinating what catches the eye and what details linger in our memory. PS: Ever look at a picture and think, "Oh, that’s just a QR code"? But what if there’s more to it? #ai #photo #technology #discoverwithimagga #Imagga #VisualAI #ComputerVision #TechInnovation #FlashbackFriday #yolo -- Stay tuned for more insights from my AI perspective by following me, Ivan. Keep up with the latest in AI by following Imagga. And remember, follow your passion, and you'll thrive!

🤖 How is Computer Vision transforming the world of art? From creating groundbreaking digital masterpieces to enhancing visitor experiences in museums and preserving cultural heritage, Computer Vision is redefining the way we engage with creativity and history. 🦾 In our latest blog post, we explore the intersection of art and technology, unveiling the possibilities that AI brings to creation, interaction, and preservation. 💡 Discover how these advancements are shaping the future of cultural experiences and opening doors to innovation like never before. Let us know your thoughts—how do you see AI influencing the art world in the years to come? 💬 - #computervision #artificialintelligence #art