5月28日交叉学科论坛学术报告二则（Lin GU & Zheng Wang National Institute

5月28日交叉学科论坛学术报告二则（Lin GU & Zheng Wang National Institute

类别：未知发布人：admin 浏览次数：次发布时间：2018-05-25 22:16

报告题目1： Medical AI under insufficient Data
报告时间： 2018年5月28日下午15:00-16:30（周一）
报告地点：计算机学院B403
报告人： Dr. Lin GU
报告人单位：National Institute of Informatics, Japan

报告人简介：
Lin GU received the Ph.D degree at the Australian National University in 2014. He is currently a Project Researcher at National Institute of Informatics (NII), Japan. He is also a visiting scholar at Kyoto University. He has published over 10 research papers in top-tier journals and conferences, including IEEE Transactions on Medical Imaging (TMI), IEEE Transactions on Image Processing (TIP), ICCV, CVPR. He has won the best paper award of ICPR 2012 and DICTA 2013. His current research interests include computational photography and medical AI.
报告摘要：
In this talk, I will focus on semi-supervised learning for biomedical image segmentation, so as to take advantage of huge unlabelled data. Observing that there usually exist some homogeneous connected areas of low confidence in biomedical images, which tend to confuse the classifier trained with limited labelled samples. To cope with this difficulty, we propose to construct forest oriented super pixels(voxels) to augment the standard random forest classifier, in which super pixels(voxels) are built upon the forest based code. Compared to the state-of-the-art, our proposed method shows superior segmentation performance on challenging 2D/3D biomedical images.

报告题目2：Scale-adaptive Low-resolution Person Re-identification

报告时间： 2018年5月28日下午16:30-18:00（周一）
报告地点：计算机学院B403
报告人： Dr. Zheng Wang
报告人单位：National Institute of Informatics, Japan

报告人简介：
Zheng Wang received the Ph.D. degree at Wuhan University in 2017. He is a Project Researcher at National Institute of Informatics (NII), Japan. He is now working under the JST CREST project "Development and Integration of Artificial Intelligence Technologies for Innovation Acceleration". He has published over 10 research papers in top-tier journals and conferences, including IEEE Transactions on Cybernetics (TCYB), IEEE Transactions on Multimedia (TMM), IJCAI, ACM Multimedia. He has won the best paper award of PCM 2014, and ACM Wuhan Doctoral Dissertation Award. His current research interests include instance search and multimedia data mining.
报告摘要：
Person re-identification is an important task in video surveillance and forensics applications. Most of previous approaches are based on a key assumption that all person images have uniform and sufficiently high resolutions. Actually, various low-resolutions and scale mismatching always exist in open world. We name this kind of problem as Scale-Adaptive Low Resolution Person Re-identification (SALR-REID). We proposed two methods for this task, i.e., (1) learning discriminating surface for scale-distance function (SDF) and (2) Cascaded Super-Resolution GAN (CSR-GAN). Extensive evaluations on two simulated datasets and one public dataset demonstrate the advantages of these two methods.

上一篇：2018年5月18日学术报告二则（吴方向 University of Saskatchewan，陈振邦国防科技大学）

下一篇：2018年 5月30日学术报告（2场）