Cornell computer scientists have discovered a way to clothe animated figures in movies or video games in higher-quality, more realistic garments, unveiling a new method for simulating complex, three-dimensional knitted fabrics.
The new process creates digital textiles that are comparable in appearance to real textiles, creating animated models of knitted clothing with “a rich variety of patterns that would be completely impractical to model using traditional techniques,” according to a study the scientists published.
Prior to the development of the new method, graphic artists’ ability to simulate a cloth-like material was limited, according to a University press release.
However, the researchers — Prof. Doug James and Prof. Steve Marschner, computer science, along with Prof. Cem Yuksel, computer science, University of Utah, and Jonathan Kaldor Ph.D ’11 — have expanded this process to work for designs such as cable-knit sweaters. The group created three-dimensional models of single stitches, then duplicated them and interwove them like tiles in a mosaic.
Each stitch is interactively looped and curved around another — much like in actual knitting — to produce 3-D shapes, according to the study.
The researchers then integrate these shapes — referred to as a “polygon mesh” — into an interactive model called a “stitch mesh.”
“We are actually changing the shape of all of the yarn loops that make up the stitches, simulating how they wrap around other loops,” Marschner said in a University press release.
Then, according to James, the stitch meshes are “relaxed” in a simulator so the digital garment fits the figure it is meant to clothe.
In the past, garments for interactive computer models have been made with triangular meshes that were of lower quality, James said.
“[This is] the first time that anybody has actually modeled these kind of knitted garments or knitted patterns in 3-D with the same kind of fidelity as real textiles,” James said.
The project was funded by the National Science Foundation, the Alfred P. Sloan Foundation, the John Simon Guggenheim Memorial Foundation and Pixar, according to the press release.