Today in Generative Media

Amazon's AI dubs; Ad industry's AI reckoning; What happens when AI hype ends?

News and Opinion

• In Further Assault on Cinema, Amazon Is Deploying AI-Aided Dubs on Streaming Movies (Futurism)

• The Ad Industry’s A.I. Reckoning (New York Times)

• The AI hype is just like the blockchain frenzy – here’s what happens when the hype dies (The Conversation)

• Chinese AI Companies Dodge U.S. Chip Curbs by Flying Suitcases of Hard Drives Abroad (Wall Street Journal)

• Are we ready to hand AI agents the keys? (MIT Technology Review)

Software

• ByteDance's Seedance 1.0 is trading blows with Google's Veo 3 (The Decoder)

• Ming-lite-omni is a unified multimodal model capable of processing images, text, audio, and video, while demonstrating strong proficiency in both speech and image generation. (HuggingFace)

• now ComfyUI has native LoRA training support. (X)

Research

• VideoMat: Extracting PBR Materials from Video Diffusion Models (project page)

• Text-Aware Image Restoration with Diffusion Models (project page)

• Large Inverse Rendering Model for Progressive Reconstruction of Shape, Materials and View-dependent Radiance Fields (project page)

• Drive Any Mesh: 4D Latent Diffusion for Mesh Deformation from Video (arXiv)

• Edit360: 2D Image Edits to 3D Assets from Any Angle (arXiv)

• TransGI: Real-Time Dynamic Global Illumination With Object-Centric Neural Transfer Model (arXiv)

• ReGuidance: A Simple Diffusion Wrapper for Boosting Sample Quality on Hard Inverse Problems (arXiv)

Misc

• An Introduction to Flow Matching and Diffusion Models (arXiv)