research
Semantic Segmentation with Dense Prediction Transformers via Grasshopper
Semantic Segmentation using Dense Prediction Transformers via Grasshopper. This innovative component leverages DPT technology to segment and identify objects within images, utilizing the ADE20K pretrained model. Inspired by the Intelligent Systems Lab at Intel Labs, this tool has been further developed within ATk. It completes the original concept of DPT tools as discussed in the ‘Coding Architecture’ book chapter.
AI Outpainting with Stable Diffusion via Grasshopper
The Ambrosinus Toolkit v1.2.8 now includes a new feature: AI outpainting. This addition enhances the Grasshopper platform with another AI tool that utilizes Stable Diffusion technology. So, it will be possible to explore your design even outside the usual on-board frame!
AI as Rendering Engine: Variation mode locally via Grasshopper
With Ambrosinus Toolkit v1.2.6 I finally brought the “Variation” mode to the locally executable Stable Diffusion-based AI component package.
This well-known mode is based on the principle of altering the initial prompt and the image provided as input. In fact, unlike Text-to-Image (T2I), Image-to-Image (I2I) always requires an initial source image.
MeshyAI toolbox for creating 3D assets now through Grasshopper
As many already know, new platforms and services dedicated to AI are growing visibly. Meshy is one of them and promises to empower content creators to effortlessly turn text and images into captivating 3D assets in under a minute. Waiting for further developments… I’m starting to integrate these features within Ambrosinus Toolkit v1.2.6 😉
