Today in Generative Media
NY mayor makes AI robocalls; Leonardo.ai raises $31M; Creativity in the age of genAI
Since When Does Eric Adams Speak Spanish, Yiddish and Mandarin? (New York Times)
Sydney-based generative AI art platform Leonardo.Ai raises $31M (TechCrunch)
Creativity in the age of generative AI: A new era of creative partnerships (TechXplore)
‘There’s no winning strategy’: the pacy, visually stunning film about the dangers of AI – made by AI (The Guardian)
Fei-Fei Li and the binders full of women in AI (VentureBeat)
Moment in Manzanar is an interactive educational film, featuring a conversational experience powered by a generative AI.
DeepCache: Accelerating Diffusion Models for Free (project page)
Style Aligned Image Generation via Shared Attention (project page)
ReconFusion: 3D Reconstruction with Diffusion Priors (project page)
FaceStudio: Put Your Face Everywhere in Seconds (project page)
X-Adapter: Adding Universal Compatibility of Plugins for Upgraded Diffusion Model (project page)
LivePhoto: Real Image Animation with Text-guided Motion Control (project page)
DPHMs: Diffusion Parametric Head Models for Depth-based Tracking (project page)
Fine-grained Controllable Video Generation via Object Appearance and Context (project page)
StableDreamer: Taming Noisy Score Distillation Sampling for Text-to-3D (arXiv)
DiffiT: Diffusion Vision Transformers for Image Generation (arXiv)
Inventions: From an ongoing series on Instagram. (Jonathan Hoefler) Click on the images to see the (fictional) descriptions!
My 70 Stable Diffusion image gens per second is old news. I'm now just under 150 per second on my home 4090. (X) StreamFusion: 100fps text-to-image. This is done on a RTX4090 locally. (X)