This weekend in Generative Media
OpenAI plays hardball vs. NYT; GenAI without lawsuits, Jared Leto invests in GenAI
News
OpenAI’s Tactics Test First Amendment in New York Times Fight (Bloomberg Law)
The Push to Develop Generative A.I. Without All the Lawsuits (New York Times)
Jared Leto Invests in Generative AI Video Startup (Hollywood Reporter)
The Data That Powers A.I. Is Disappearing Fast (New York Times)
IATSE Ratifies New Three-Year Deal, Despite AI Worries (Variety)
Figma explains how its AI tool ripped off Apple’s design (The Verge)
Former Google employees got $16 million to make video games more fun with AI (Quartz)
OpenAI’s latest model will block the ‘ignore all previous instructions’ loophole (The Verge)
Software
Arbitrary-Scale Video Super-Resolution with Structural and Textural Priors (ECCV2024) (GitHub)
fofr / ultrapixel: Ultra high resolution images (up to 4096x4096) based on Stable Cascade (Replicate)
Research
Streetscapes: Large-scale Consistent Street View Generation Using Autoregressive Video Diffusion (SIGGRAPH 2024, project page)
Lite2Relight: 3D-aware Single Image Portrait Relighting (SIGGRAPH 2024, project page)
4D-Rotor Gaussian Splatting: Towards Efficient Novel View Synthesis for Dynamic Scenes (SIGGRAPH 2024, ACM)
I2V-Adapter: A General Image-to-Video Adapter for Diffusion Models (SIGGRAPH 2024, ACM)
Universal Facial Encoding of Codec Avatars from VR Headsets (SIGGRAPH 2024, arXiv)
IDOL: Unified Dual-Modal Latent Diffusion for Human-Centric Joint Video-Depth Generation (ECCV 2024, project page)
VGGSfM: Visual Geometry Grounded Deep Structure From Motion (ECCV 2024, project page)
AccDiffusion: An Accurate Method for Higher-Resolution Image Generation (project page)
LogoSticker: Inserting Logos into Diffusion Models for Customized Generation (project page)
KineTy: Kinetic Typography Diffusion Model (project page)