Today in Generative Media

Axios changes strategy vs. AI; Gig workers write for AI; Mistral aims at US tech giants

• Axios Sees A.I. Coming, and Shifts Its Strategy (New York Times) “The premium for people who can tell you things you do not know will only grow in importance, and no machine will do that,” says Jim VandeHei, C.E.O. of Axios.

• Gig workers are writing essays for AI to learn from (Business Insider)

• Europe’s A.I. ‘Champion’ Sets Sights on Tech Giants in U.S. (New York Times)

• New AI music generator Udio synthesizes realistic music on demand (Ars Technica)

• The AI Filmmaking Mega-Party. April 15, 2024 at 8PM. Las Vegas. (Curious Refuge)

• STORM is a research prototype for automating the knowledge curation process. (Stanford Open Virtual Assistant Lab)

• Lixel CyberColor: Next-Gen 3D Content Technology Based on Multi-SLAM & 3DGS (XGrids)

• Online demo for `Bilateral Reference for High-Resolution Dichotomous Image Segmentation` (HuggingFace)

• InstantMesh: Efficient 3D Mesh Generation from a Single Image with Sparse-view Large Reconstruction Models (GitHub)

• mixtral-8x22b-v0.1-4bit (Replicate)

• MuRF: Multi-Baseline Radiance Fields (CVPR 2024, project page)

• GoMAvatar: Efficient Animatable Human Modeling from Monocular Video Using Gaussians-on-Mesh (CVPR 2024, project page)

• Reducing the Memory Footprint of 3D Gaussian Splatting (I3D 2024, project page)

• ControlNet++: Improving Conditional Controls with Efficient Consistency Feedback (project page)

• AWOL: Analysis WithOut synthesis using Language (arXiv)

• We just merged Grounding DINO in the Transformers library. It's a popular model for zero-shot object detection (no labels) with 500+ citations already! It can work in tandem with SAM or Stable Diffusion to replace specific things in an image. (X)

• There is a new Canvas feature on RenderNet… In this video below, I show how user-friendly it is to move around images to create a storyboard for the protagonist in a story. (X)

• C3PO x Childish Gambino 🤖 👑 100% AI (X)