This demo is intended to be a full walkthrough of the process from creating an ML model, and using this ML model directly in Houdini. But it is also built with exploration in mind. Plug in your own terrain erosion HDAs, or heightfield patterns. Use higher or lower resolution to compare results. Test against massive terrain, or work with small fine tune details. Train a moon crater ML model, or alien world. Or simply just use existing real world lidar data.
The ONNX Machine Learning Terrain demo comes in two parts.
Part One | Training: (ONNX_teraining_Terrain.hip) Walk through a simple setup of how to leverage PDG, and a handful of terrain HDAs processors, to create hundreds or thousands of unique terrain assets. Feeding those terrain files into COPs to export image pairs which will be used by the TrainPix2Pix HDA, which will create a custom ONNX file.
Part Two | Inference: (ONNX_Paint_Terrain.hip) Use the newly created ONNX file(or use the one provided), with Houdini 20's new ONNX SOP. With the hip files basic heightfield set up, edit its shape and design with a Paint Texture Mask directly in the viewport, and watch the ML model generate a final detailed heightfield.