Hi Elliott, This is a great breakdown and very generous in sharing your process and insights, you came a long way from the vending machine days!
Are you planning on releasing the UE4 project to the public? Or only builds? I'd love to play around with it in the editor if possible!
Most recently procedural technology became one of the central parts of various popular solutions for 3d artists. It’s being used extensively in SideFx Houdini and Substance Designer from Allegorithmic, giving an opportunity to achieve incredible realism in 3d. There are other examples of implementing this tech in 3d design tools. One of them is the amazing “Interactive Sketching of Urban Procedural Models” project by Ignacio Garcia-Dorado.
Ignacio Garcia-Dorado is an engineer, who’d currently working in Computational Imagining team at Google Research. He’s got a Ph.D. in Purdue University and his main focus is procedural modeling particularly in the field of 3d urban reconstruction. Basically, “Interactive Sketching of Urban Procedural Models” is the secret weapon for all urban planners, who want to create great 3d visualisations, without working too much with the 3d modeling software. Though we believe this technology will also be invaluable to the game development community.
3D modeling remains a notoriously difficult task for novices despite significant research effort to provide intuitive and automated systems. We tackle this problem by combining the strengths of two popular domains: sketch-based modeling and procedural modeling. On the one hand, sketch-based modeling exploits our ability to draw but requires detailed, unambiguous drawings to achieve com- plex models. On the other hand, procedural modeling automates the creation of precise and detailed geometry but requires the tedious definition and parameterisation of procedural models. Our system uses a collection of simple procedural grammars, called snippets, as building blocks to turn sketches into realistic 3D models. We use a machine learning approach to solve the inverse problem of finding the procedural model that best explains a user sketch. We use non-photorealistic rendering to generate artificial data for training convolutional neural networks capable of quickly recognizing the procedural rule intended by a sketch and estimating its parameters. We integrate our algorithm in a coarse-to-fine urban modeling system that allows users to create rich buildings by successively sketch- ing the building mass, roof, facades, windows, and ornaments. A user study shows that by using our approach non-expert users can generate complex buildings in just a few minutes.
Ignacio Garcia-Dorado, Google
So far this project is not finished and you can’t get this software to run on your PC, unfortunately. We really hope to hear more about Ignacio’s work in the upcoming days during SIGGRAPH 2016. Imagine what kind of worlds can you build with it!