Today in Generative Media

Nikon fights back; Artists will train their own models; Faster diffusion on iOS & MacOS

• Nikon Fights Back Against AI Images, Touts ‘Natural Intelligence’ (PetaPixel)

• Artists will train their own models/styles/generators. Users will be able to generate style-consistent art from these models. (Twitter)

• Faster Stable Diffusion with Core ML on iPhone, iPad, and Mac (HuggingFace)

• ICCV 2023 Workshop on AI for Creative Video Editing and Understanding - October 2, 2023, Paris, France.

• TryOnDiffusion: A Tale of Two UNets (arXiv)

• DreamHuman: Animatable 3D Avatars from Text (project page)

• VidEdit: Zero-Shot and Spatially Aware Text-Driven Video Editing (project page)