国产精品搬运,欧美最猛黑人xxxx,亚洲日本va午夜在线电影,国产午夜视频在线观看

首頁
聯系我們
您現在所在的位置:首頁  首頁  學術活動
張健教授學術報告會

報告題目:High Density Crowd People Counting in Computer Visionblob

報告人:Associate Professor Jian Zhang

報告時間:20171025   上午9:00-11:00

報告地點:綜合樓710會議室

邀請人:機器人科學與工程學院  吳成東  教授

Abstract

Counting pedestrians and measuring crowd density play an essential role for crowd monitoring applications including physical security, public space management, and retail space design. For the popular density-map-estimation based counting framework, there are mainly two questions involved for accurate counting result: i). accurate recognition of target object and background; ii). Precise estimation of density values of target objects. This talk first begins with a literature review of crowd counting algorithms: including traditional hand-crafted feature based methods to the newly emerging deep learning based crowd counting methods, and then we present our work towards the two fatal problems mentioned above.

For more precise density estimation under the challenging situation of drastic object scale variations, we present a novel deep learning framework for crowd counting by learning a perspective-embedded deconvolution network. Through layer-wise fusion, perspective pyramid is merged into the deconvolution network, driving the network to learn to combine the underlying scene geometric constraints adaptively and obtain more accurate, perspective-aware density values. 

For accurately identify target object areas from clutter background, deep network is usually the first choice. However, for dense density crowd counting, plain deep network does not guarantee the counting performance. This is due to the natural conflict between the very limited size of objects and the large down sampling stride with many pooling layers in the deep network. To address this problem, we introduce a novel cascade architecture for crowd counting. Several shallow networks are cascaded and counting results are refined subsequently through each stage. We also propose a grid-based loss function to introduce regional supervisions for more accurate density value estimation.

Short Bio:

Dr. Jian Zhang received the BSc. degree from East China Normal University, Shanghai, China, in 1982; the MSc. degree in computer science from Flinders University, Adelaide, Australia, in 1994; and the Ph.D. degree in electrical engineering from the University of New South Wales (UNSW), Sydney, Australia, in 1999.

From 1997 to 2003, he was with the Visual Information Processing Laboratory, Motorola Labs, Sydney, as a Senior Research Engineer, and later became a Principal Research Engineer and a Foundation Manager with the Visual Communications Research Team. From 2004 to July 2011, he was a Principal Researcher and a Project Leader with National ICT Australia, Sydney, and a Conjoint Associate Professor with the School of Computer Science and Engineering, UNSW. He is currently an Associate Professor with the Advanced Analytics Institute, School of software, Faculty of engineering and Information Technology, University of Technology Sydney, Sydney. Prof Zhang’s research interests include multimedia signal processing, computer vision, pattern recognition, visual information mining, human-computer interaction and intelligent video surveillance systems. Prof Zhang has co-authored more than 130 paper publications, book chapters, patents and technical reports from his research output, he was the co-author of eight granted US and China patents.

 Dr. Zhang is an IEEE Senior Member. He was Technical Program Chair, 2008 IEEE Multimedia Signal Processing Workshop; Associated Editor, IEEE Transactions on Multimedia; Associated Editor, IEEE Transactions on Circuits and Systems for Video Technology (T-CSVT) ; Associated Editor, EURASIP Journal on Image and Video Processing. Dr Zhang was Guest Editor of T-CSVT for Video Technology for Special Issue (March 2007) of the Convergence of Knowledge Engineering Semantics and Signal Processing in Audiovisual Information Retrieval. As a General Co-Chair, Jian chaired the International Conference on Multimedia and Expo (ICME 2012) in Melbourne Australia 2012. As a Technical Program Co-Chair, Jian chaired The IEEE Visual Communications and Image Processing (IEEE VCIP 2014).


  • <sup id="efjtm"><dfn id="efjtm"><ul id="efjtm"></ul></dfn></sup>
  • <ruby id="efjtm"><div id="efjtm"><label id="efjtm"></label></div></ruby>
    <abbr id="efjtm"></abbr>

      <ruby id="efjtm"></ruby>
      <sup id="efjtm"><dfn id="efjtm"><ul id="efjtm"></ul></dfn></sup>