Improving probes in dynamic diffuse global illumination
| Thesis title in Czech: | Vylepšení sond v metodě pro dynamické rozptýlené globální osvětlení |
|---|---|
| Thesis title in English: | Improving probes in dynamic diffuse global illumination |
| Key words: | globální osvětlení|vykreslování v reálném čase|nepřímé osvětlení |
| English key words: | global illumination|real-time rendering|indirect lighting |
| Academic year of topic announcement: | 2021/2022 |
| Thesis type: | diploma thesis |
| Thesis language: | angličtina |
| Department: | Department of Software and Computer Science Education (32-KSVI) |
| Supervisor: | Mgr. Tomáš Iser, Ph.D. |
| Author: | hidden - assigned and confirmed by the Study Dept. |
| Date of registration: | 26.11.2021 |
| Date of assignment: | 26.11.2021 |
| Confirmed by Study dept. on: | 04.01.2022 |
| Date and time of defence: | 08.02.2022 10:00 |
| Date of electronic submission: | 07.01.2022 |
| Date of submission of printed version: | 10.01.2022 |
| Date of proceeded defence: | 08.02.2022 |
| Opponents: | Mgr. Martin Kahoun |
| Guidelines |
| Global illumination is one of the fundamental problems in computer graphics. In the real world, we perceive objects around us because they are illuminated by various light sources, such as the sun or artificial light bulbs. However, only a small portion is typically illuminated directly by photons that travel straight from the light source towards the objects. The rest is illuminated indirectly, which means that the light bounced (reflected) from other objects first. This, for example, explains why we can see objects in a room that only has a small window, because the small amount of sunlight gets reflected from the white walls and illuminates the room indirectly. We call this global illumination, and while we can relatively easily solve this problem in offline Monte Carlo rendering, it still poses a significant challenge in real-time applications.
The aim of this thesis is to investigate the current approaches for solving the global illumination problem in real-time applications, e.g., computer games or interactive architectural visualizations. Then, one of the state-of-the-art solutions will be picked and implemented in an existing real-time rendering engine. This will allow analyzing the behaviour of such a solution in a dynamic custom scene that requires global illumination to be rendered, e.g., an indoor scene with moving objects inside and is only illuminated by sunlight from tiny windows. The goal is to evaluate how well the solution behaves, identify problems that the solution has, and suggest, implement, and evaluate possible improvements. Specifically, Majercik et al. [2019] recently presented a dynamic diffuse global illumination (DDGI) method. As this method is based on probes in a grid, it also suffers from all the related problems, e.g., preventing probes from being placed inside dynamic geometry or avoiding artefacts related to probe updates. |
| References |
| 1) Zander Majercik, Jean-Philippe Guertin, Derek Nowrouzezahrai, and Morgan McGuire. Dynamic diffuse global illumination with ray-traced irradiance fields. Journal of Computer Graphics Techniques (JCGT), 8(2):1–30, June 2019. ISSN 2331-7418. URL http://jcgt.org/published/0008/02/01/.
2) Philip Dutré, Kavita Bala, and Philippe Bekaert. Advanced Global Illumination, Second Edition. Taylor & Francis, 2003. URL http://www.advancedglobalillumination.com/. |