关于举办华南理工大学“海内外优秀青年学者论坛”计算机科学与工程学院分论坛的通知
日期:2017年05月11日 09:25:18

广大师生:

华南理工大学“海内外优秀青年学者论坛”旨在面向全球邀请拥有不同学术背景的青年才俊,围绕国际科学前沿、热点研究领域以及行业产业的技术问题等展开探讨和交流。希望藉此平台,互相启迪、开拓视野,增强国际交流与合作。本次论坛的具体安排如下:

1. 论坛时间

2017年5月11日下午15:00-16:30

2. 论坛地点

广州大学城华南理工大学计算机学院B3-213学术报告厅

3. 论坛议程

(1) 15:00-15:10 学院领导致欢迎辞

(2) 15:10-15:50 报告人:王君乐 博士,研发主管,Research and Development Division, Ars Nova Systems, France

报告题目:From 2D to stereoscopic 3D visual attention–Psychophysical method and computational modeling

(3) 15:50-16:30 报告人:李静 博士,University of Nantes, France

报告题目:Subjective assessment methodology for multi–dimensional visual experience

欢迎广大师生参加!

 

 

计算机科学与工程学院

2017年5月10日

 

报告1:From 2D to stereoscopic 3D visual attention – Psychophysical method and computational modeling

报告摘要:

Visual attention is an important mechanism deployed in the human visual system to reduce the amount of information that our brain needs to process. An increasing amount of efforts are being dedicated in the studies of visual attention, and many computational models performing well in predicting salient areas of 2D images have been proposed. Recently, the emerging applications of stereoscopic 3D display and VR bring additional depth information affecting the human viewing behavior, and require extensions of the efforts made in 2D visual attention modeling. In this talk, a computational model of visual attention for stereoscopic 3D still image is introduced. Apart from detecting salient areas based on 2D visual features, the proposed model takes depth as an additional visual dimension. The measure of depth saliency is derived from the eye movement data obtained from an eye-tracking experiment using synthetic stimuli.

报告人简介:

Junle Wang received his Ph.D. degree in 2012 from University of Nantes, France. He then became an assistant professor with the Department of Electronic and Digital Technologies, University of Nantes, France and a visiting research fellow with Centre for Multimedia and Network Technology, Nanyang Technological University, Singapore. He is currently in charge of the Research and Development Division, Ars Nova Systems, France. He serves as a regular reviewer for international journals and conferences, including TIP, TNNLS, SIGGRAPH. His research interests include image classification, visual saliency detection, deep learning, image quality assessment, human visual perception, depth perception, eye-tracking, and psychophysical experimentation.

 

报告2:Subjective assessment methodology for multi-dimensional visual experience

报告摘要:

Multimedia technology is aiming to improve people’s viewing experience, seeking for better immersiveness and naturalness. The development of HDTV, 3DTV, Ultra HDTV and VR are recent illustrative examples of this trend. The Quality of Experience (QoE) in multimedia encompass multiple perceptual dimensions. For instance, in 3DTV and VR, three primary dimensions have been identified: image quality, depth quality and visual comfort. In this talk, focusing on the most advanced multimedia technology, i.e., 3DTV (including VR) and Ultra HDTV, one basic question about QoE is studied: “how to subjectively assess QoE taking care of its multidimensional aspect?” This talk will show you why the traditional standardized 2D subjective quality assessment methods do not work anymore in this condition, why Pair Comparison is considered as the most reliable method and how to improve its efficiency.

报告人简介:

Jing Li received her Ph.D. degree in 2013 in the Image and Video Communications Group (IVC) in Polytech Nantes, University of Nantes, France. Her research interests are subjective assessment methodologies on Quality of Experience (QoE), visual discomfort and Preference of Experience (PoE) in 3DTV and Ultra HDTV. She has been an assistant professor in University of Nantes since 2014. Her proposed subjective assessment methodologies for QoE, namely ARD and ORD, have been adopted by the ITU standard and IEEE P3333.1 standard. She has published more than 20 papers in international conferences and journals. She serves as a regular reviewer for more than 10 conferences and journals. She is also the group leader in VQEG for the construction of 3D QoE Ground Truth database, and responsible for the cooperation of VQEG and DVB for 3DTV test plan.