Bidirectional Path Tracing II CS295, Spring 2017 Shuang Zhao - PowerPoint PPT Presentation

Bidirectional Path Tracing II CS295, Spring 2017 Shuang Zhao Computer Science Department University of California, Irvine CS295, Spring 2017 Shuang Zhao 1

Last Lecture • Bidirectional path tracing CS295, Spring 2017 Shuang Zhao 2

Today’s Lecture • Bidirectional path tracing II CS295, Spring 2017 Shuang Zhao 3

Recap: Building Paths in Batches • Then, connecting each pair of vertices y s , z t yields a full path sample • ( n L n E ) full path samples constructed from one pair of light and sensor sub-paths CS295, Spring 2017 Shuang Zhao 4

Recap: Path Contribution • Given a light sub-path with n L vertices and a sensor sub-path wit n E vertices, each resulting path sample with has contribution: where CS295, Spring 2017 Shuang Zhao 5

Special Case: s = 0, 1 No visibility term in Simple path tracing Path tracing with next-event estimation CS295, Spring 2017 Shuang Zhao 6

Special Case: t = 0, 1 No visibility term in Simple adjoint particle tracing Adjoint particle tracing with next-event estimation CS295, Spring 2017 Shuang Zhao 7

Survival Probability and Albedo • Assuming Reflectance albedo then implies and CS295, Spring 2017 Shuang Zhao 8

Image Construction D D(u, v) • The value I j for pixel j equals • D is the image plane (parameterized using u , v ) • • h j is the filter function (e.g., 2D Gaussian) • I ( u , v ) is the image function CS295, Spring 2017 Shuang Zhao 9

Example Scene CS295, Spring 2017 Shuang Zhao 10

Individual Components CS295, Spring 2017 Shuang Zhao 11

Individual Components (Weighted) CS295, Spring 2017 Shuang Zhao 12

Next Lecture • Refraction & Bidirectional Scattering Functions • Radiative transfer CS295, Spring 2017 Shuang Zhao 13