LogIn
Two new projects have been added to SideFX's Content Library, a collection of scene files and Houdini Digital Assets designed for study or integration into your work. The PCA Dejitter and ML Sketching are now available for download, along with accompanying tutorials offering a more detailed explanation by Jakob Ringler.
Sometimes, we encounter jittery geometry due to faulty simulations or external capture processes, and there may not be enough time or the option to resim, recapture, or fix the issue properly. In such cases, Principal Component Analysis (PCA) can be used to de-jitter the geometry. This setup leverages PCA's dimensionality reduction features to eliminate unwanted movements while preserving the overall shape of the geometry.
ML Sketching showcases how to use the new ML SOP/TOP nodes to create data generation and training pipelines within Houdini without writing any code.
The main idea is to predict geometry generation based on user drawings. Instead of estimating actual geometry, the neural network predicts the parameter values used in a procedural generator. This approach keeps the setup lightweight and flexible, allowing it to be adapted to work with various procedural generators that create geometry with a clear silhouette.
You can grab these projects here and explore more new entries:
Also, don't forget to join our 80 Level Talent platform and our new Discord server, follow us on Instagram, Twitter, LinkedIn, Telegram, TikTok, and Threads, where we share breakdowns, the latest news, awesome artworks, and more.