Today in Generative Media

Deepfakes and fraud losses, Meta lawsuit; AI "promptography"; Zoom's AI rebrand

News

• Billionaire Wins Court Battle Against Meta Over His Image in Deepfake Ads (PetaPixel)

• Deepfakes of Elon Musk contribute to billions in fraud losses (Yahoo)

• These artists are rebranding AI image generation with a new name (Fast Company)

• Zoom 2.0 relaunches as an AI-first company without video in its name (The Verge)

• Tech start-up and AI publisher Spines wants to release 8000 books in a year (AV Club)

• Riot Games Calls Netflix Using AI for Arcane Image "Disrespectful" to Artists (80 Level)

Software

• Train Flux LoRAs For Portraits: fal-ai/flux-lora-portrait-trainer (FAL)

• PlayAI clones voices on command (TechCrunch)

• Luma expands Dream Machine AI video model into full creative platform, mobile app (VentureBeat)

• Ai2’s open source Tülu 3 lets anyone play the AI post-training game (TechCrunch)

• llama-cpp-python-gradio is a Python package that makes it easy for developers to create machine learning apps powered by llama.cpp models using Gradio. (GitHub)

• Smol TTS keeps getting better! Introducing OuteTTS v0.2 - 500M parameters, multilingual with voice cloning! … Model weights on the hub, you can even run this on a Raspberry Pi! (X)

• Diffusers' pipelines can be used as an inference engine for a server. It supports concurrent and multithreaded requests to generate images that may be requested by multiple users at the same time. (GitHub)

Research

• Quark: Real-time, High-resolution, and General Neural View Synthesis (SIGGRAPH Asia 2024 project page)

• MotionFix: Text-Driven 3D Human Motion Editing (SIGGRAPH Asia 2024 project page)

• MyTimeMachine: Personalized Facial Age Transformation (project page)

• Material Anything: Generating Materials for Any 3D Object via Diffusion (project page)

• Dynamics-Aware Gaussian Splatting Streaming Towards Fast On-the-Fly Training for 4D Reconstruction (project page)

• ZeroGS: Training 3D Gaussian Splatting from Unposed Images (project page)

• SplatFlow: Multi-View Rectified Flow Model for 3D Gaussian Splatting Synthesis (project page)

• Fugatto 1: Foundational Generative Audio Transformer Opus 1 (project page)

Misc

• CS492(D): Diffusion Models and Their Applications (KAIST) [Lecture slides and recordings]

• Jacob Collier x Gen Music | Google Lab Sessions | Engineer Cut (YouTube)