A few weeks in Generative Media

Hollywood to own 3D scans forever; AP & OpenAI sharing; Google's AI journalism tool

I’m back! I still found time to clip a handful of articles while I was away:

• Hollywood Studios Want to Scan Background Actors for One Day's Pay, Use Likenesses for 'Rest of Eternity,' SAG Claims (People)

• AP strikes news-sharing and tech deal with OpenAI (Axios)

• Google Tests A.I. Tool That Is Able to Write News Articles (New York Times)

• Meta claims its new art-generating model is best-in-class (TechCrunch)

• Generative AI’s secret sauce — data scraping— comes under attack (VentureBeat)

• DreamGenerator! 📸✨ A camera that transforms your photos into something new using the power of generative AI. (Twitter)

• TokenFlow: Consistent Diffusion Features for Consistent Video Editing (project page)

• DiffDreamer: Towards Consistent Unsupervised Single-view Scene Extrapolation with Conditional Diffusion Models (ICCV 2023, project page)

• To Infinity and Beyond: SHOW-1 and Showrunner Agents in Multi-Agent Simulations (project page). “Create episodes of TV shows with a prompt - SHOW-1 will write, animate, direct, voice, edit for you.”

• SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis (arXiv)

• Animate-A-Story: Storytelling with Retrieval-Augmented Video Generation (arXiv)