Zhibo Wang

SenseTime, China

Affiliations: SenseTime.
Address: Beijing, China.
Email: wangzhibo@sensetime.com.

王至博 (Zhibo Wang) is currently a Research Scientist at SenseTime, ARFace Group. He received his Ph.D. degree in Software Engineering from School of Software, Tsinghua University, in 2022. He received his B.S degree in Microelectronics Science and Technology from Nanjing University, Nanjing, China, in 2017. His research interests include facial animation, face reconstruction and facial image editing.

news

Feb 28, 2023	Our paper “Learning a Morphable Face Reflectance Model from Low-cost Data” got accepted by CVPR 2023.
Feb 28, 2023	Our paper “ShadowNeuS: Neural SDF Reconstruction by Shadow Ray Supervision” got accepted by CVPR 2023.
Jul 2, 2022	Our paper “Structure-aware Editable Morphable Model for 3D Facial Detail Animation and Manipulation” got accepted by ECCV 2022.
May 17, 2022	I successfully defended my Ph.D. dissertation!.
Apr 12, 2022	Our paper “Semantically Disentangled Variational Autoencoder for Modeling 3D Facial Details” got accepted by TVCG.

selected publications

Single Image Portrait Relighting via Explicit Multiple Reflectance Channel Modeling

Zhibo Wang, Xin Yu, Ming Lu, Quan Wang, Chen Qian, and Feng Xu

ACM Transactions on Graphics 2020

Abs HTML PDF Supp Video Website

Portrait relighting aims to render a face image under different lighting conditions. Existing methods do not explicitly consider some challenging lighting effects such as specular and shadow, and thus may fail in handling extreme lighting conditions. In this paper, we propose a novel framework that explicitly models multiple reflectance channels for single image portrait relighting, including the facial albedo, geometry as well as two lighting effects, i.e., specular and shadow. These channels are finally composed to generate the relit results via deep neural networks. Current datasets do not support learning such multiple reflectance channel modeling. Therefore, we present a large-scale dataset with the ground-truths of the channels, enabling us to train the deep neural networks in a supervised manner. Furthermore, we develop a novel module named Lighting guided Feature Modulation (LFM). In contrast to existing methods which simply incorporate the given lighting in the bottleneck of a network, LFM fuses the lighting by layer-wise feature modulation to deliver more convincing results. Extensive experiments demonstrate that our proposed method achieves better results and is able to generate challenging lighting effects.