Today in Generative Media
Creators paid for AI training videos; MSFT files suit over AI use; AI junk science
News
AI Companies Are Paying Content Creators For Their Unpublished Videos (PetaPixel)
Microsoft sues service for creating illicit content with its AI platform (Ars Technica)
AI-fabricated 'junk science' floods Google scholar, researchers warn (phys.org)
'The New York Times' takes OpenAI to court. ChatGPT's future could be on the line (NPR)
Biden Administration Adopts Rules to Guide A.I.’s Global Spread (New York Times)
OpenAI Courts Trump With Vision for ‘A.I. in America’ (New York Times)
Raspberry AI raises $24M from a16z to accelerate fashion design (TechCrunch)
Software
The Complete AI Toolkit For Unreal Engine Developers (Ludus AI)
Kokoro is a frontier TTS model for its size of 82 million parameters (text in/audio out). (HuggingFace)
Run ComfyUI workflows for free with Gradio on Hugging Face Spaces (HuggingFace)
Luma AI - Introducing Ray2 (YouTube) [Stream begins at 10am PST, 1/15/24]
Research
Seaweed APT Diffusion Adversarial Post-Training for One-Step Video Generation (project page)
Arc2Avatar: Generating Expressive 3D Avatars from a Single Image via ID Guidance (project page)
UCo3D: UnCommon Objects in 3D (project page) [dataset]
DI-PCG: Diffusion-based Efficient Inverse Procedural Content Generation for High-quality 3D Asset Creation (project page)
MatchAnything: Universal Cross-Modality Image Matching with Large-Scale Pre-Training (project page)
SynShot - Synthetic Prior for Few-Shot Drivable Head Avatar Inversion (project page)
Sky-T1: Train your own O1 preview model within $450 (project page)
MEt3R: Measuring Multi-View Consistency in Generated Images (project page)
AI Agents Are Here. What Now? (HuggingFace)
CamCtrl3D: Single-Image Scene Exploration with Precise 3D Camera Control (arXiv)
3DGS-to-PC: Convert a 3D Gaussian Splatting Scene into a Dense Point Cloud or Mesh (arXiv)