GAMES Webinar 2018-48期（CVPR 2018论文）| 周晓巍（浙江大学），吴潜溢（中国科学技术大学），王宗伟（上海科技大学）
【GAMES Webinar 2018-48期（CVPR 2018论文）】
报告时间：2018年5月31日（星期四）晚20:00 – 20:30（北京时间）
报告题目：Learning to Estimate 3D Human Pose and Shape from 2D Image
Recovering 3D human pose is a challenging problem with many applications. It has been conditionally solved by motion capture systems or depth sensors. This talk will discuss the more challenging case of using a single RGB camera: going directly from 2D appearance to 3D geometry. While deep learning approaches have shown remarkable abilities to solve 2D vision problems, it is difficult for them to directly learn and predict 3D geometry due to the lack of training data and higher dimensionality and nonlinearity of the solution space. In this talk, I will introduce our recent efforts toward 3D human pose and shape prediction from a single image, which solve the aforementioned challenges by integrating deep learning with geometric models as well as end-to-end learning using weakly-annotated data and multi-view geometry.
周晓巍，浙江大学计算机学院CAD&CG国家重点实验室研究员、国家青年千人计划入选者。2008年获浙江大学信息工程专业学士学位，2013年获香港科技大学电子及计算机工程博士学位。2014年至2017年在美国宾夕法尼亚大学计算机及信息科学系、GRASP机器人实验室从事博士后研究。主要研究领域为计算机视觉、机器人感知、医学图像分析，尤其在三维物体识别、人体姿态估计、图像匹配等方面取得了一系列重要成果。策划和组织了Geometry Meets Deep Learning Workshops，并长期担任PAMI、IJCV、TIP等二十余种SCI期刊审稿人以及CVPR、ICCV、IJCAI等计算机领域顶级会议程序委员会委员。 个人主页：https://fling.seas.upenn.edu/~xiaowz/dynamic/wordpress/
报告时间：2018年5月31日（星期四）晚20:30 – 21:00（北京时间）
报告题目：Alive Caricature from 2D to 3D
Caricature is an art form that expresses subjects in abstract, simple and exaggerated views. While many caricatures are 2D images, this talk presents an algorithm for creating expressive 3D caricatures from 2D caricature images with minimum user interaction. The key idea of our approach is to introduce an intrinsic deformation representation that has the capability of extrapolation, enabling us to create a deformation space from standard face datasets, which maintains face constraints and meanwhile is sufficiently large for producing exaggerated face models. Built upon the proposed deformation representation, an optimization model is formulated to find the 3D caricature that captures the style of the 2D caricature image automatically. The experiments show that our approach has better capability in expressing caricatures than those fitting approaches directly using classical parametric face models such as 3DMM and FaceWareHouse. Moreover, our approach is based on standard face datasets and avoids constructing complicated 3D caricature training sets, which provides great flexibility in real applications.
Qianyi Wu is currently an Master student in School of Mathematical Sciences at University of Science and Technology of China (USTC), under the supervision of Assoc Prof. Juyong Zhang. He obtained his Bachelor’s degree from USTC in 2016. His research interests include computer vision and computer graphics, especially 3D face modeling.
报告时间：2018年5月31日（星期四）晚21:00 – 21:30（北京时间）
报告题目：Face Aging with Identity-Preserved Conditional Generative Adversarial Networks
Face aging is of great importance for cross-age recognition and entertainment related applications. However, the lack of labeled faces of the same person across a long age range makes it challenging. Because of different aging speed of different persons, our face aging approach aims at synthesizing a face whose target age lies in some given age group instead of synthesizing a face with a certain age. By grouping faces with target age together, the objective of face aging is equivalent to transferring aging patterns of faces within the target age group to the face whose aged face is to be synthesized. Meanwhile, the synthesized face should have the same identity with the input face. Thus we propose an Identity-Preserved Conditional Generative Adversarial Networks (IPCGANs) framework, in which a Conditional Generative Adversarial Networks module functions as generating a face that looks realistic and is with the target age, an identity-preserved module preserves the identity information and an age classifier forces the generated face with the target age. Both qualitative and quantitative experiments show that our method can generate more realistic faces in terms of image quality, person identity and age consistency with human observations.
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；