GAMES Webinar 2018-56期(Siggraph 2018论文)| 徐泽祥(加州大学圣迭戈分校),胡渊鸣(麻省理工学院)



【GAMES Webinar 2018-56期(Siggraph 2018论文)】
报告嘉宾1:徐泽祥,加州大学圣迭戈分校
报告时间:2018年7月19日(星期四)晚8:00 – 8:45(北京时间)
主持人:董悦,微软亚洲研究院(个人主页:http://yuedong.shading.me/
报告题目:Deep Image-Based Relighting from Optimal Sparse Samples
报告摘要:
We present an image-based relighting method that can synthesize scene appearance under novel, distant illumination from the visible hemisphere, from only five images captured under pre-defined directional lights. Our method uses a deep convolutional neural network to regress the relit image from these five images; this relighting network is trained on a large synthetic dataset comprised of procedurally generated shapes with real-world reflectances. We show that by combining a custom-designed sampling network with the relighting network, we can jointly learn both the optimal input light directions and the relighting function. We present an extensive evaluation of our network, including an empirical analysis of reconstruction quality, optimal lighting configurations for different scenarios, and alternative network architectures. We demonstrate, on both synthetic and real scenes, that our method is able to reproduce complex, high-frequency lighting effects like specularities and cast shadows, and outperforms other image-based relighting methods that require an order of magnitude more images.
讲者简介:
Zexiang Xu is currently a third year Ph.D. student in computer science at University of California, San Diego, advised by Prof. Ravi Ramamoorthi. His research interests are about computer graphics and computer vision, including rendering, appearance and geometry capture, hair modeling and light fields.
讲者个人主页:http://cseweb.ucsd.edu/~zex014/

 

报告嘉宾2:胡渊鸣,麻省理工学院
报告时间:2018年7月19日(星期四)晚8:45 – 9:30(北京时间)
主持人:董悦,微软亚洲研究院(个人主页:http://yuedong.shading.me/
报告题目:Exposure:基于增援学习和对抗生成网络实现无穷分辨率白箱图片处理
从Tensorflow到Taichi:我们需要一个什么样的计算机图形学库?
报告摘要:
基于可微照片编辑模型,增援学习(Reinforcement Learning)和生成对抗网络(Generative Adversarial Network, GAN),我们的Exposure系统实现了无穷分辨率的图片处理,使得GAN在高质量照片处理领域得以实用化。同时,Exposure能够生成人类可理解的操作序列,以达到逆向工程艺术风格的目的。
正如很多深度学习项目一样,Exposure是用TensorFlow开发的。遗憾的是,在计算机图形学中我们还没有一个像TensorFlow一样的工具。为了缓解计算机图形学中面向研究的工具链缺乏的问题,我主导开发了taichi。我会分享开发过程中的一些有趣的发现和挑战,包括关于可移植性、性能、可扩展性、易用性的取舍,以及这些设计决策如何影响团队协作等。希望这些讨论能够在搭建通用计算机图形学库这一工程中起到抛砖引玉的作用。
讲者简介:
胡渊鸣,麻省理工学院一年级博士生,师从Wojciech Matusik。2017年本科毕业于清华大学姚班(优秀毕业生),期间访问东京大学、斯坦福大学、微软亚洲研究院、宾夕法尼亚大学,获得ACM-ICPC / NOI / APIO金牌四枚,MIT Edwin S. Webster奖学金等奖项。研究兴趣为基于物理的动画和渲染、计算摄影学等,在SIGGRAPH / TOG / CVPR / SCA发表一作论文四篇,主导开发开源计算机图形学库taichi。参与IEEE TIP / ACM SIGGRAPH Asia审稿工作。
讲者个人主页:http://taichi.graphics/me

 

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;
GAMES主页的“资源分享”有往届的直播讲座的视频及PPT等。

 

观看直播的链接:http://webinar.games-cn.org/

You may also like...