This weekend in Generative Media

Apple plans AI deals with publishers; AI ethics falls by the wayside; Midjourney V6

• Apple Explores A.I. Deals With News Publishers (New York Times)

• This week in AI: AI ethics keeps falling by the wayside (TechCrunch)

• Midjourney V6 is here with in-image text and completely overhauled prompting (VentureBeat)

• Midjourney will "find you and collect that money" if you infringe any IP with v6 (Decoder)

• An artist fights back, and Midjourney has embarrassed themselves (Gary Marcus on Substack)

• Creators, porn stars turn to AI doppelgangers to keep fans entertained (Washington Post)

• Apple wants AI to run directly on its hardware instead of in the cloud (Ars Technica)

• The Simulation by Fable open sources AI tool to power Westworlds of the future (VentureBeat)

• Journalists Had 'No Idea' About OpenAI's Deal to Use Their Stories (Wired)

• America’s Big AI Safety Plan Faces a Budget Crunch (Wired)

• A ‘thirsty’ generative AI boom poses a growing problem for Big Tech (CNBC)

• Wizards of the Coast doubles down on generative AI stance, says artists are ‘what makes D&D great’ (GeekWire)

• Longtime gaming leader Jordan Weisman on his new AI-driven game development platform (GeekWire)

• An AI Haunted World (Ethan Mollick on Substack)

• AI-Generated Sci-Fi Novel Secures Prestigious Literary Prize in China (Medium)

• The Move One iOS App makes it easy to capture and create 3D animations with just a phone (move.ai)

• After working on my AI assisted free 2.5D point and click adventure game for a year, I am releasing a demo for PC and OS X! This demo shows how AI tools help a solo dev make a game and what it could be. I have the whole process in my blog (X) An experimental freeware 2.5D adventure game featuring AI-assisted graphics (Echoes of Somewhere)

• Align Your Gaussians: Text-to-4D with Dynamic 3D Gaussians and Composed Diffusion Models (project page)

• DreamTuner: Single Image is Enough for Subject Driven Generation (project page)

• Splatter Image: Ultra-Fast Single-View 3D Reconstruction (project page)

• pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (project page)

• Repaint123: Fast and High-quality One Image to 3D Generation with Progressive Controllable 2D Repainting (project page)

• Paint3D: Paint Anything 3D with Lighting-less Texture Diffusion Models (project page)

• Intrinsic Image Diffusion for Single-view Material Estimation (project page)

• Zero-Shot Metric Depth with a Field-of-View Conditioned Diffusion Model (project page)

• Global Latent Neural Rendering (project page)

• Fairy: Fast Parallelized Instruction-Guided Video-to-Video Synthesis (project page)

• VidToMe: Video Token Merging (project page)

• Adaptive Guidance: Training-free Acceleration of Conditional Diffusion Models (arXiv)

• Analyzing and Improving the Training Dynamics of Diffusion Models (arXiv)

• StreamDiffusion is an innovative diffusion pipeline designed for real-time interactive generation. It introduces significant performance enhancements to current diffusion-based image generation techniques (GitHub)

• All-In-One Music Structure Analyzer (GitHub) Check out this excellent demo!

• OpenLRM is an open-source implementation of Large Reconstruction Models.

  Image-to-3D in 10 seconds! (HuggingFace)

• Exploring Architectural Fundamentals at 60 FPS (LinkedIn)

• I made a script that watches for new screenshots and renames them using Llava 13B: (X)

• I’m excited to share a fun little 🎁 today: an *in-browser* AI image generator for 340x340 pixel art and SoTA open source diffuser models. (X)

• We're hosting a CVPR workshop on AI-assisted art---a big focus is to understand how AI models are currently being used in artistic workflows (to help inspire the next generation of better, more useful AI tools). (X)

• Made a movie using Midjourney V6 + RunwayML and Live Action. The time is now for AI to push our creative at any level. (X)