Course Content Improvement, British Style

Separating Style and Content

In many vision problems, we want to infer two (or more) hidden factors which interact to produce our observations. We may want to disentangle illuminant and object colors in color constancy; rendering conditions from surface shape in shape-from-shading; face identity and head pose in face recognition; or font and letter class in character recognition. We refer to these two factors generically a...

Indirect Rule: French and British Style

Content Aware Neural Style Transfer

In the emerging inter-disciplinary field of art and image processing, algorithms have been developed to assist the analysis of art work. In most applications, especially brush stroke analysis, high resolution digital images of paintings are required to capture subtle patterns and details in the high frequency range of the spectrum. Algorithms have been developed to learn styles of painters from...

Separating Style and Content for Generalized Style Transfer

Neural style transfer has drawn broad attention in recent years. However, most existing methods aim to explicitly model the transformation between different styles, and the learned model is thus not generalizable to new styles. We here attempt to separate the representations for styles and contents, and propose a generalized style transfer network consisting of style encoder, content encoder, m...

Course Content and Goals

The basic question: what can and cannot be computed? ˆ Hilbert’s 10th problem, 1900. (CANNOT) ˆ Error testing (a.k.a. the Halting Problem). (CANNOT) ˆ Godel’s Theorem. (CANNOT) Tools: ˆ Well specified models. ˆ Mathematical arguments (convincing and unassailable) — analogous to geometry in high school. Important goal of the course: ˆ Have you better able to recognize when an argument is mathema...