Representing scenes with multiple semi-transparent colored layers has be...
Existing image generator networks rely heavily on spatial convolutions a...
Modeling daytime changes in high resolution photographs, e.g., re-render...
Representation learning becomes especially important for complex systems...
Despite recent advances in computer vision based on various convolutiona...