[Photo: SIGGRAPH Asia 2008]

Research Intern

Affiliation: School of Computing, National University of Singapore

Addr: School of Computing, National University of Singapore
Computing 1 13 Computing Drive Singapore 117417 Republic of Singapore

Email: luozhipi AT gmail DOT com (prefer), luozhipi AT comp.nus.edu.sg

MSN: helios1688@hotmail.com

Birth date: Mar. 13. 1986

Hometown: Shenzhen, Guangdong, China


Publications Projects Demo Experience Skills

 

*News

Online DEMO of ACM Multimedia 2009 is available: ViewFocus

NUS-WIDE image dataset is released: NUS-WIDE

About me

I am a third year Master student with B.Eng degree of South China University of Technology(SCUT), meanwhile I am a research intern of LMS (Lab for Media Search), NUS (National University of Singapore), advised by Prof. Chua Tat-Seng.

RESEARCH INTERESTS-What can we benefit from Internet multimedia mining?
  1. We face challenges of large-scale context-aware image/video retrieval. Parallel computing eara? Scalable indexing and distribution systems?

  2. Which views of world's places can we mine from the vast amount noisy Internet multimedia? Beyond Google Earth by providing multi-view browsing of landmarks.

  3. How will we mine patterns for machine learning ways? The issues about generating featuers.
PUBLICATIONS[Back to Top]
Zhiping Luo, Haojie Li, Jinhui Tang, Richang Hong and Tat-Seng Chua. "ViewFocus: Explore Places on Google Maps Using Photos with View Direction Filtering''. ACM Multimedia 2009 (MM'09). (PDF,Demo).
Zhiping Luo, Haojie Li, Jinhui Tang, Richang Hong and Tat-Seng Chua. "Estimating Poses of World's Photos with Geographic Metadata''. The 16th International Conference on MultiMedia Modeling (MMM), Chongqing, China. (PDF)
Tat-Seng Chua, Jinhui Tang, Richang Hong, Haojie Li,Zhiping Luo,and Yan-Tao Zheng. "NUS-WIDE: A Real-World Web Image Database from National University of Singapore''. ACM International Conference on Image and Video Retrieval (CIVR'09). Greece. Jul. 8-10, 2009.(Project, PDF).
Richang Hong, Jinhui Tang, Zheng-Jun Zha,Zhiping Luo, and Tat-Seng Chua. "Mediapedia: Mining Web Knowledge to Construct Multimedia Encyclopedia''. The 16th International Conference on MultiMedia Modeling (MMM), Chongqing, China. (PDF)
Yanling Wu, Mei Wang, Guangda Li, Zhiping Luo, Tat-Seng Chua, Xumin Liu. "VDictionary: Automatically Generate Visual Dictionary via Wikimedias''. The 16th International Conference on MultiMedia Modeling (MMM), Chongqing, China. demo. (PDF)
Rong-Hua Li, Zhiping Luo, Guoqiang Han. "Pseudo-inverse Locality Preserving Projections''. IEEE International Conference on Computational Intelligence and Security 2009(CIS'09). Beijing, China (matlab, PDF)
Zhiping Luo, Xingming Zhang. "A Semi-Supervised Learning Based Relevance Feedback Algorithm in Content-Based Image Retrieval''. Chinese Conference on Pattern Recognition (CCPR 2008). Beijing, China. Oct. 22-24,2008.(IEEE Explore).
Chun-bo Liu, Hui-jin Wang, Zhi-ping Luo, Xiu-qin Yu, Li-hua Liu. "QoS Multicast Routing Problem Based on Artificial Fish-Swarm Algorithm''. First International Workshop on Education Technology and Computer Science, 2009 (ETCS'09). Volume 2, Issue , 7-8 March 2009 Page(s):814 - 817. (IEEE Explore).
Zhiping Luo, Jin-tian Su, Rong-Hua Li, Xing-Ming Zhang, Zhe Xiao. "Design and implementation of WAP 2.0-based campus portal'', journal of Application Research and Computers. Vol.26 (2), 2009. (in Chinese) (CEPS, PDF).
RESEARCH EXPERIENCE[Back to Top]

Sep.2008-Present, National University of Singapore, Research Intern.

referee: Prof. Chua Tat-Seng, Prof. Chong-Wah Ngo

  1. Participated in the NUS-WIDE project to develop the image annotation tool for overall 26W Flickr images' labeling. This tool, namely NUS-LabelMe, aims to reduce the labor efforts for image annotation. Developed a modified active learning algorithm to be implanted in that tool and then the tool can automatically find the top semantic relevant images out of unlabeled images. To find those images only needs few images annotated for a concept.

  2. Worked out several methods to process the overall 26W Flickr images associated with text tags. Those methods include stop word removal using WordNet, spelling check using Google Search, compute Google Distance using Google Image Search, approximate nearest neighbors (ANN) algorithm for training with six kinds of image features.

  3. Proposed a method to find the precise photos with respect to a place by view direction filtering on Google Maps. Developed a novel system named ViewFocus based on the method. This system lets users view world's places with more precise photos.

  4. Worked out several dimensionality reduction algorithms to adaress the high-dimensional, undersampled problems in fields of data mining, machine learning and computer vision. Different to others, ours can effectively see thenonlinear manifolds in the raw data space, and preserve the data structure in the reduced space well.

Sep.2007-Sep.2008, South China University of Technology, Research Assistant

referee: Prof. Zhang Xiang-Ming, Jianfu Chen (SUNYSB)

  1. Worked on Face Recognition and Video Surveillance. Proposed LDA+SVM(Linear Discriminant Analysis) algorithm for face recognition. Developed algorithms of image translation on mobile. Participated in the video surveillance project of Zhongshan city, partner is China Mobile, Zhongshan city.
  2. Leader of the project "WAP 2.0 portal of Guangzhou Higher Education Mega Center", partner is China Mobile, Guangzhou. city.

Sep. 2007-Sep. 2008, SCUT Microsoft Technique Club, Leader of club's official website

  1. Leading developer for the forum module of the club's website.
  2. Core member of club's ACM algorithm team.

Sep. 2006-Jul. 2007, ZTE Software Co., Ltd, Joint Laboratory Trainee

  1. Developer of the GPS SmartNavigator project.
PROJECTS[Back to Top]

Web-Scale Multimedia Corpus Construction

  1. We created a web image dataset, the dataset includes 269,648 images and the associated tags from Flickr, with a total number of 5018 unique tags; six-low-level features extracted from these images; ground-truth for 81 concepts that can be used for evaluation; a baseline results for web image annotation by learning the tags using the traditional k-NN algorithm.

WAP 2.0 portal of Guangzhou Higher Education Mega Center, this work is supported by National 973 Project (G20000263), P.R.China

  1. We designed and implemented a WAP portal of Guangzhou higher education mega center based on WAP 2.0 technique. This portal provides online education and campus information for students. More specially, we mainly worked on the adaptive image delivery in wireless network model.
Programing Skills and Corresponding Qualifications[Back to Top]

Programming

  1. Languange: C++, C#, JAVA, Python, Matlab.
  2. OS: UNIX, WINDOWS.
  3. Others: course on clustering computing (MPI).

Contest Awards

  1. Web 2.0 resource discovering and storage platform: Third Award of student projects, supported by Microsoft Research Asia, 2007.

  2. Mobile BusLine system: Excellent Award of student projects, supported by Microsoft Research Asia, 2007.

  3. Mobile information retrieval system: Winners of the 2nd innovation contest in Guangzhou Higher-Education Town, supported by Microsoft Research Asia & SCUT, 2007.

  4. Mobile campus portal: The 2nd prize for the first contest of mobile application software, supported by HSBC & SCUT, 2007.

Qualifications:

  1. Qualification Certificate of Software Designer; Qualification level: middle.

  2. Qualification Certificate of System Analyst; Qualification level: senior.

* I have passed national examination organized by the Chinese government department and have obtained qualifications for Computer and Software Technology Proficiency. These certificates are licensed both in P.R.China and Japan.

Links
Locations of visitors to this page