Today in Generative Media

Ted Sarandos & George Lucas on AI; AI in indie games; Sony interview snafu

• Netflix’s Ted Sarandos On A.I. Threat To Hollywood: “A.I. Is Not Going To Take Your Job, The Person Who Uses A.I. Well Might” (Deadline)

• George Lucas Thinks Artificial Intelligence in Filmmaking Is 'Inevitable' (IGN)

• Brain Drain: A.I. And Indies (Game Informer)

• Sony apologizes for Sony interview with Sony developer Neil Druckmann (PC Gamer)

• A Devil’s Bargain With OpenAI (The Atlantic)

• How Generative AI Will Change The Jobs Of Artists And Designers (Forbes)

• Marigold Pipelines for Computer Vision Tasks (HuggingFace)

• ToonCrafter: Generative Cartoon Interpolation (GitHub) 🤗 ToonCrafter can interpolate two cartoon images by leveraging the pre-trained image-to-video diffusion priors. Please check our project page and paper for more information.

• See More Details: Efficient Image Super-Resolution by Experts Mining (HuggingFace)

• MusePose: a Pose-Driven Image-to-Video Framework for Virtual Human Generation (GitHub)

• DreamMat: High-quality PBR Material Generation with Geometry- and Light-aware Diffusion Models (SIGGRAPH 2024, project page)

• MoSca: Dynamic Gaussian Fusion from Casual Videos via 4D Motion Scaffolds (project page)

• Collaborative Video Diffusion: Consistent Multi-video Generation with Camera Control (project page)

• GFlow: Recovering 4D World from Monocular Video (project page)

• DGD: Dynamic 3D Gaussians Distillation (project page)

• NPGA: Neural Parametric Gaussian Avatars (project page)

• I2VEdit: First-Frame-Guided Video Editing via Image-to-Video Diffusion Models (project page)

• PCM : Phased Consistency Model (project page)

• Matryoshka Multimodal Models (project page)

• 3DitScene: Editing Any Scene via Language-guided Disentangled Gaussian Splatting (project page)

• EM Distillation for One-step Diffusion Models (arXiv)

• 3D Reconstruction with Fast Dipole Sums (arXiv)

• DiffCalib: Reformulating Monocular Camera Calibration as Diffusion-Based Dense Incident Map Generation (arXiv)

• Splat-SLAM: Globally Optimized RGB-only SLAM with 3D Gaussians (arXiv)

• Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability (arXiv)

• NVS-Solver: Video Diffusion Model as Zero-Shot Novel View Synthesizer (arXiv)

• Here's how you can render any 3D scene in any style and have full control over the final image by creating individual prompts for all the objects in your scene. 👇 (X)