GAMES Webinar 2022 – 216期(隐式表征：从重建到生成) | Michael Niemeyer (Max Planck Insitute for Intelligent Systems)
【GAMES Webinar 2022-216期】(视觉专题-隐式表征：从重建到生成，Talk+Panel形式)
报告嘉宾：Michael Niemeyer(Max Planck Insitute for Intelligent Systems)
报告题目：Implicit Neural Scene Representations and 3D-Aware Generative Modelling(CVPR 2021 best paper)
Deep generative models allow for photorealistic image synthesis at high resolutions. But for many applications, this is not enough: content creation also needs to be controllable. While several recent works investigate how to disentangle underlying factors of variation in the data, most of them operate in 2D and hence ignore that our world is three-dimensional. Further, only few works consider the compositional nature of scenes. Our key hypothesis is that incorporating a compositional 3D scene representation into the generative model leads to more controllable image synthesis. Representing scenes as compositional generative neural feature fields allows us to disentangle one or multiple objects from the background as well as individual objects’ shapes and appearances while learning from unstructured and unposed image collections without any additional supervision. Combining this scene representation with a neural rendering pipeline yields a fast and realistic image synthesis model. As evidenced by our experiments, our model is able to disentangle individual objects and allows for translating and rotating them in the scene as well as changing the camera pose. This talk starts by discussing our works on implicit 3D representations and then move to our newest projects GRAF and GIRAFFE, and will briefly touch on our newest work CAMPARI.
I’m Michael, a third-year PhD student in the field of computer vision and machine learning in the Autonomous Vision Group (AVG) in Tübingen, Germany. Before starting my PhD in late 2018, I studied Mathematics in Cologne, Germany and for my Master’s in Computer Science, I went to Scotland to study at the St. Andrews University. My research focuses on 3D representations for deep neural networks. I’m a big fan of implicit or coordinate-based 3D representations, and I’m interested in how they can be inferred from sparse observations.
廖依伊，浙江大学特聘研究员。分别于西安交通大学和浙江大学获得学士与博士学位。2018至2021年，她在德国马克斯普朗克智能系统研究所及图宾根大学从事博士后研究，师从CVPR PAMI 青年研究员奖得主Andreas Geiger教授。她的研究兴趣为三维视觉，包括场景重建、场景语义理解、可控图像生成等，作为第一负责人搭建了包含全面大规模语义及样例标签的数据集KITTI-360。近年来发表论文二十篇，包括TIP、CVPR、ICCV、NeurIPS、ICRA等十余篇顶尖期刊/会议论文。担任BMVC 2021领域主席。个人主页：https://yiyiliao.github.io/。
崔兆鹏，浙江大学计算机科学与技术学院CAD&CG国家重点实验室“百人计划”研究员、博士生导师。2017年在加拿大西蒙弗雷泽大学获得博士学位，2017年至2020年在瑞士苏黎世联邦理工学院计算机视觉和几何实验室任高级研究员。研究方向为三维计算机视觉，具体包括三维重建、运动恢复结构、多视角立体几何、三维场景理解、同时定位与地图构建、视频图像编辑等。近年来在计算机视觉、机器人、计算机图形学、机器学习等领域的顶级期刊和会议上发表论文30余篇，曾获ICRA 2020机器视觉最佳论文提名和IROS 2021安全、安保和救援机器人最佳论文提名。
Lingjie Liu is Lise Meitner Postdoctoral Research Fellow working with Prof. Christian Theobalt in the Visual Computing and AI Department at the Max Planck Institute for Informatics. She received her Ph.D. degree at the University of Hong Kong in 2019. Before that, she got her B.Sc. degree in Computer Science at Huazhong University of Science and Technology in 2014. Her research interests include neural scene representations, neural rendering, human performance modeling and capture, and 3D reconstruction. Webpage: https://lingjie0206.github.io/.
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；