Overview

Description

CALENDAR DESCRIPTION:

Computational Photography is concerned with overcoming the limitations of traditional photography with computation: in optics, sensors, and geometry; and even in composition, style, and human interfaces. The course covers computational techniques to improve the way we process, manipulate, and interact with visual media. The covered topics include image-based lighting and rendering, camera geometry and optics, computational apertures, advanced image filtering operations, high-dynamic range, image blending, texture synthesis and inpainting. Students with credit for CMPT 451 may not take this course for further credit.

COURSE DETAILS:

This is a complex course consisting of multiple stages. In the first 2 weeks, we have an overview of fundamental image processing followed by fundamental computational photography topics in mathematical modeling and also in real-world film production environments through guest lectures. The latter part of the course follows a research-heavy curriculum. 2 hours of each week is dedicated to deep dives to fundamental topics in image manipulation including edit propagation and soft segmentation, and computational photography including monocular depth estimation and intrinsic decomposition, relating multiple papers on each topic with each other in terms of mathematical modeling. 1 hour each week is reserved for all-together open-ended discussions on term projects and research papers. Students form project groups and each projects develops their own photography project through discussions with the instructor. Each student also prepares a detailed video presentation of a selected research paper, which is watched and discussed together in class at the end of the semester. Flipped classroom setup in overview weeks allows for covering any open questions of students on fundamental topics, and also discussing the research field in general, introducing coding tools for photography etc. This is done in an open environment where every student participates in the discussion. In the research-focused lectures, we go over very detailed formulations directly from pdf's of papers. In the discussion hours, the students casually talk about their plans for their project or bring up problems they came across. The students define and create unique applications in project groups. The collaboration is enhanced through weekly check-ins and discussions during lecture hours.

Grading

Materials

REQUIRED READING NOTES: