Meta Platforms, Inc. has launched AssetGen 2.0, an advanced generative AI foundation model designed to create high-quality 3D assets from text and image prompts. This new version builds on the original AssetGen by introducing two specialized models: one for generating 3D meshes and another for producing textures. AssetGen 2.0 enables the transformation of simple text, image, and even world prompts into production-ready 3D meshes and textures, significantly improving the fidelity and quality of generated 3D content. The technology aims to accelerate the 3D asset creation process, facilitating faster transition from concept to production, particularly for Meta's Horizon platform. Additionally, the update includes enhancements such as a new AI texturing pipeline that supports high-quality physically based rendering (PBR) materials, automated advanced retopology, and UV map optimization. Meta describes AssetGen 2.0 as a substantial technical advancement compared to its predecessor, targeting higher quality and fidelity in 3D asset generation.
So @AIatMeta announced a new 3D asset generator called Meta 3D AssetGen 2.0. Meta describes AssetGen 2.0 as a “significant leap forward” compared to AssetGen 1.0, aiming for much higher quality and fidelity in generated 3D assets through a different technical approach, with the
עבודה מאוד מרשימה של מטא: מודל שיודע לייצר אובייקטים תלת מידיים מטקסט בלבד. יש להם מאוחרי הקלעים גם מודל שאחראי באופן ספציפי על מרקמים וטקסטורות. https://t.co/VfU6VpMXFi
A big update of Live AI extension for @RoamResearch has been released ! Including: - Image generation - Mermaid & argument map generation - Web search with OpenAI or Anthropic models ... and a lot of other improvements and fixes. https://t.co/HAnireanmI