Today in Generative Media

AI action figures; News as a two-way conversation; Korea bans AI song copyrights

News and Opinion

• A.I. Action Figures Flood Social Media (Accessories Included) (New York Times)

• The next big AI shift in media? Turning news into a two-way conversation (Fast Company)

• Korean Copyright Association Bans AI-Composed Songs From Copyright Registration (Billboard)

• Paying artists for AI art would be 'cool,' says OpenAI CEO Sam Altman (The Decoder)

• Why Adobe Firefly might be the only AI image tool that actually matters (ZDNet)

• ChatGPT became the most downloaded app globally in March (TechCrunch)

Software

• YouTube rolls out a free AI music-making tool for creators (TechCrunch)

• Introducing Pika Twists: manipulate any character or object in your footage, while keeping the rest perfectly intact. (X)

• ComfyUI-Veo2-Experimental A custom node extension for ComfyUI that integrates Google's Veo 2 text-to-video generation capabilities. (GitHub)

• Pusa: Thousands Timesteps Video Diffusion Model (GitHub)

Research

• Comprehensive Relighting: Generalizable and Consistent Monocular Human Relighting and Harmonization (project page)

• Geo4D: Leveraging Video Generators for Geometric 4D Scene Reconstruction (project page)

• Sonic: Shifting Focus to Global Audio Perception in Portrait Animation (project page)

• HoloPart: Generative 3D Part Amodal Segmentation (project page)

• In-2-4D: Inbetweening from Two Single-View Images to 4D Generation (project page)

• Generative AI for Film Creation: A Survey of Recent Advances (arXiv)

• Parametric Shadow Control for Portrait Generation in Text-to-Image Diffusion Models (arXiv)

Misc

• Synthesia just crossed a huge milestone – $100M in ARR + a strategic investment from Adobe. (LinkedIn)

• This object removal workflow kinda slaps… It’s a Wan 2.1 workflow that’s using Flux Fill to inpaint the first frame. (X)

• A short film program to explore AI on screen (Google Keyword blog)