20160504-13 Shenlong Wang: 3D Scene Understanding through Maps of the World

2016-4-29 21:03| 发布者: 程一-计算所| 查看: 7608| 评论: 0

摘要: 报告嘉宾2：Shenlong Wang (University of Toronto, Canada)报告时间：2016年5月4日（星期三）晚21:00（北京时间）报告题目：Lost Shopping! Monocular Localization in Large Indoor Spaces主持人：左旺孟（哈工大 ...

报告嘉宾2：Shenlong Wang (University of Toronto, Canada)

报告时间：2016年5月4日（星期三）晚21:00（北京时间）

报告题目：3D Scene Understanding through Maps of the World

主持人：左旺孟（哈工大）

报告摘要：Map information provides rich semantic and geometric knowledge about the specific scene captured in each individual image, even beyond the image’s field of view. This is crucial to resolve visual ambiguities and improve the performance of scene understanding. In the first part of the talk, we attempt to investigate the utilization of geo-location information for outdoor holistic scene understanding. In the second project, we propose a novel single image localization algorithm for large indoor spaces using a shopping mall map.

参考文献：

[1] Holistic 3D Scene Understanding from a Single Geo-tagged Image Shenlong Wang, Sanja Fidler and Raquel Urtasun International Conference on Computer Vision and Pattern Recognition (CVPR), Boston, 2015 (Oral)

[2] Lost Shopping! Monocular Localization in Large Indoor Spaces Shenlong Wang, Sanja Fidler and Raquel Urtasun International Conference on Computer Vision (ICCV), Santiago, 2015 (Oral)

报告人简介：

Shenlong Wang is a PhD student in Machine Learning Group at Department of Computer Science, University of Toronto since Jan 2014. His advisor is Prof. Raquel Urtasun. Prior to that, he spent one semester at Toyota Technological Institute at Chicago. He received his master degree in pattern recognition and bachelor degree in computational science, both from Northwestern Polytechnical University (2006-2013). During 2011-2013, he worked at The Hong Kong Polytechnic University as a research assistant under the supervision of Prof. Lei Zhang. He worked at National Laboratory of Pattern Recognition, Chinese Academy of Sciences as a research intern in Prof. Stan Li's group from 2009-2010. His general research interest lies in computer vision and machine learning. He has published several top conference papers such as ICCV, CVPR, NIPS, etc.

报告材料[Slides]

收藏邀请

上一篇：20160504-13 马超: Hierarchical Convolutional Features for Visual Tracking下一篇：20160511-14 潘金山:Blind Image Deblurring Using Dark Channel Prior

20160504-13 Shenlong Wang: 3D Scene Understanding through Maps of the World

最新评论

相关分类