GAMES Webinar 2020 – 132期(绘制专题) | Cheng Zhang(UC Irvine), Yuchi Huo(KAIST)

【GAMES Webinar 2020-132期】(绘制专题)

报告嘉宾1:Cheng Zhang, UC Irvine


报告题目:Differentiable Rendering Theory and Applications


Physics-based differential rendering, the process of computing the derivatives of radiometric measurements with respect to scene parameters, opened the door to a number of challenging inverse problems including material and geometry reconstruction from photographs. Unfortunately, computing derivatives of radiometric measurements remains challenging in the presence of complex light transport effects and nontrivial scene geometry. We introduce a differential theory of radiative transfer, which shows how individual components of the radiative transfer equation (RTE) can be differentiated with respect to arbitrary differentiable changes of a scene. To numerically estimate the derivatives given by our theory, we introduce an unbiased Monte-Carlo estimator supporting arbitrary surface and volumetric configurations. We demonstrate the practical usefulness of our technique by showing a few synthetic examples inspired by real-world applications in inverse rendering, non-line-of-sight (NLOS), biomedical imaging and design.


Cheng Zhang is a Ph.D. candidate (third-year) in Computer Science department at UC Irvine, whose advisor is Prof. Shuang Zhao. Cheng’s research interest is in the area of physics-based rendering with focus on modeling and simulating of light interactions with complex material. Some of his research is also related to solving differentiable rendering and inverse rendering problem. Cheng obtained his B.E. in Electrical Engineering from BJUT (2011-2015) and his M.S. in Computer Science from Columbia University (2015-2017) under the supervision of Prof. Changxi Zheng.


报告嘉宾2:Yuchi Huo, KAIST


报告题目:Adaptive Incident Radiance Field Sampling and Reconstruction Using Deep Reinforcement Learning


Serious noise affects the rendering of global illumination using Monte Carlo (MC) path tracing when insufficient samples are used. The two common solutions to this problem are filtering noisy inputs to generate smooth but biased results and sampling the MC integrand with a carefully crafted probability distribution function (PDF) to produce unbiased results. Both solutions benefit from an efficient incident radiance field sampling and reconstruction algorithm. This study proposes a method for training quality and reconstruction networks (Q- and R-networks, respectively) with a massive offline dataset for the adaptive sampling and reconstruction of first-bounce incident radiance fields. The convolutional neural network (CNN)-based R-network reconstructs the incident radiance field in a 4D space, whereas the deep reinforcement learning (DRL)-based Q-network predicts and guides the adaptive sampling process. The approach is verified by comparing it with state-of-the-art unbiased path guiding methods and filtering methods. Results demonstrate improvements for unbiased path guiding and competitive performance in biased applications, including filtering and irradiance caching.


Graduated from CAD&CG, ZJU and working in KAIST right now.




GAMES主页的“使用教程”中有 “如何观看GAMES Webinar直播?”及“如何加入GAMES微信群?”的信息;

You may also like...