计算多媒体实验室

厦门大学信息学院   计算机系

邮件地址:

• 研究兴趣
• 开源项目
• 发表论文
• 教学经历

•     赵万磊于2002年7月和2005年11月先后在云南大学获得工学学士和工学硕士学位。在硕士期间，以联合培养的方式在中国科学院软件所学习一年。博士就读于香港城市大学，师从杨宗华教授。他于2010年11月获得哲学博士学位。自2011年2月至2013年12月先后在德国凯泽斯劳滕-兰道工业大学以及法国国家信息与自动化所（大西洋-布列塔尼分所）从事博士后研究。

      赵万磊主要从事多媒体信息检索以及多维时间序列分析方面的研究。他研究的多媒体信息检索主要聚焦于向量数据库索引结构设计，多媒体信息（文本，图像以及视觉实例）的统一特征表示。多维时间序列分析主要聚焦于在线的异常检测及多维时间序列的短程趋势预测。此外，他本人还从事优化算法方面的研究，帮助企业对生产问题进行优化建模。他的科研团队欢迎有志于将来从事科研以及投身IT行业的青年才俊加入。

目前，课题组有两个博士招生名额（2026年9月入学）两个硕士招生名额（2027年9月入学），请符合报考条件且感兴趣的同学与赵老师联系。

• 多媒体信息检索
• 计算机视觉
• 机器学习

• Traditional local interest point detectors and descriptors LIP-VIREO
• k-means and its variants xkmeans
• Indexing graph construction and graph-based fast NN search TSDG
• Product quantizer for fast NN search product-quantizer
• Mean shift for image shading and segmentation Mean-shift

• Near-duplicate Image/Video Detection
  1. Fast tracking of near-duplicate keyframes in broadcast domain with transitivity propagation, ACM Conf. on Multimedia, 2006, the 2nd author
  2. Scale-rotation invariant pattern entropy for keypoint-based near-duplicate detection, IEEE Trans. on Image Processing, 2009, the 1st author
  3. Large-scale near-duplicate web video search: Challenge and opportunity, IEEE ICME, 2009, the 1st author
  4. On the annotation of web videos by efficient near-duplicate search, IEEE Trans. on Multimeida, 2010, the 1st author
  5. Flip-Invariant SIFT for Copy and Object Detection, IEEE Trans. on Image Processing, 2013, the 1st author
  6. Near-duplicate keyframe identification with interest point matching and pattern learning, IEEE Trans. on Multimedia, 2007, the 1st author
  7. Oriented pooling for dense and non-dense rotation-invariant features, BMVC, 2013, the 1st author
  8. Sim-Min-Hash: An efficient matching technique for linking large image collections, ACM Multimedia, 2013, the 1st author
• Machine Learnig
  9. k-means: A revisit, Neurocomputing, 2018, the 1st author
  10. A joint model for IT operation series prediction and anomaly detection, Neurocomputing, 2021, the 2nd author*
  11. k-sums clustering: a stochastic optimization approach, ACM CIKM, 2021, the 1st author*
  12. Fast k-Means Based on k-NN Graph, IEEE ICDE, 2018, the 2nd author*
  13. Online Deep Metric Learning via Mutual Distillation, IEEE ICME, 2022, the 2nd author*
• Fast Indexing
  14. Approximate k-NN Graph Construction: A Generic Online Approach, IEEE Trans. on Multimeida, 2021, the 1st author*
  15. On the merge of k-NN graph, IEEE Trans. on Big Data, 2021, the 1st author*
  16. Fast k-NN Graph Construction by GPU based NN-Descent, ACM CIKM, 2021, the 2nd author*
• Computer Vision
  17. Towards Accurate Localization by Instance Search, ACM Conf. on Multimedia, 2021, the 3rd author*
  18. Deeply Activated Salient Region for Instance Search, ACM TOMM, 2022, the 2nd author*
  19. Instance search based on weakly supervised feature learning, Neurocomputing, 2020, the 2nd author*
  20. Dynamic sampling for deep metric learning, Pattern Recognition Letters, 2021, the 2nd author*
  21. Instance search via instance level segmentation and feature representation, Journal of Visual Communication and Image Representation, 2021, the 2nd author*

             The full publication list can be found from his Google Scholar website

• Multimedia Information Retrieval
• C Programming
• Optimization Theory, Method and Practice