Today in Generative Media

Perplexity: publishers don't own facts; AI safety experts urge care; Meta devours data

News and Opinion

• Perplexity AI CEO Believes No Publisher Should Own the Right to Reported Facts (Observer)

• AI safety advocates tell founders to slow down (TechCrunch)

• Meta's AI feasts on user data (Axios)

• As generative AI gets better, what will happen to artists? (TechCrunch)

• Exclusive: Ziff Davis study says AI firms rely on publisher data to train models (Axios)

• The Results of the NFS AI vs. Human Screenwriting Challenge (NoFilmSchool)

• How Film Schools Are Bracing for Change in a Time of Hollywood Disruption (TheWrap)

Software

• Motion blending is here! Blend any motion from our platform - whether it's from video or text - with seamless, AI-driven transitions. From bold moves to subtle shifts, your animations have never looked smoother! (X) [from Meshcapade]

• SuperSplat 1.8.0 is here: Open source 3D Gaussian Splat Editor (X)

• Training-free Regional Prompting for Diffusion Transformers (GitHub)

• ConsiStory: Training-Free Consistent Text-to-Image Generation [SIGGRAPH 2024] (GitHub)

• FLUX.1-dev LoRA Outfit Generator Gradio Demo (GitHub)

• This runs entirely in your browser. It uses tensorflow.js/face-api.js to detect smiling, here's the top-level code (X)

Research

• Tencent Hunyuan3D-1.0: A Unified Framework for Text-to-3D and Image-to-3D Generation (project page)

• ReMatching Dynamic Reconstruction Flow (project page)

• GenXD: Generating Any 3D and 4D Scenes (project page)

• Adaptive Caching for Faster Video Generation with Diffusion Transformers (project page)

• FewViewGS: Gaussian Splatting with Few View Matching and Multi-stage Training (NeurIPS 2024, arXiv)

• GameGen-X: Interactive Open-world Game Video Generation (arXiv)