This weekend in Generative Media
GenAI image energy use; AI toys and educational TV; AI image copyrights in China
Generating AI Images Uses as Much Energy as Charging Your Phone, Study Finds (Gizmodo)
Opinion: I made an AI pal at the Toy Fair, but I don’t want to invite him home (Washington Post)
PBS KIDS embraces AI to launch new era of children's educational TV (Mashable)
Chinese Court’s Landmark Ruling: AI Images Can be Copyrighted (AI Business)
The copyright case against AI art generators just got stronger with more artists and evidence (VentureBeat)
How to use Google Bard and ChatGPT to accelerate your creative writing (GeekyGadgets)
Law secretly drafted by ChatGPT makes it onto the books (The Register)
The power of Stable Diffusion XL in your device, offline (Noiselith)
EraseID: AI face anonymizer for advertising and privacy protection (piktid.com)
One-step Diffusion with Distribution Matching Distillation (project page)
SyncTalk: The Devil😈 is in the Synchronization for Talking Head Synthesis (project page)
Gaussian Shell Maps for Efficient 3D Human Generation (project page)
Control4D: Efficient 4D Portrait Editing with Text (project page)
Animate124: Animating One Image to 4D Dynamic Scene (project page)
Rethinking Directional Integration in Neural Radiance Fields (arXiv)
DreamSync: Aligning Text-to-Image Generation with Image Understanding Feedback (arXiv)
Using the new IPadapter batch unfold settings to rotoscope/animate a high motion fight scene! (X)