Presentation

Material acquisition using deep learning
Contributor
Event Type
ACM SIGGRAPH Asia Thesis Fast Forward
Doctoral Consortium





TimeSunday, 17 November 20199:00 - 11:00
LocationPlaza Meeting Room P2
DescriptionTexture, highlights, and shading are some of many visual cues that allow humans to perceive material appearance in pictures. Designing algorithms able to leverage these cues to recover spatially-varying bi-directional reflectance distribution functions (SVBRDFs) from a few images has challenged computer graphics researchers for decades. I explore the use of deep learning to tackle lightweight appearance capture and make sense of these visual cues.
We introduce several innovations on training data acquisition and network design, bringing clear improvement over the state of the art for lightweight material capture using as little as one picture and up to 10 images.
We introduce several innovations on training data acquisition and network design, bringing clear improvement over the state of the art for lightweight material capture using as little as one picture and up to 10 images.