Our goal is to build a web-scale content based image retrieval system. We are addressing two major challenges by harnessing the distributed computing power in MSR-SVC: 1) large scale machine learning for image representation and , 2) efficient image indexing and query. The following are several ongoing projects within WISE.
Content-Based Image Retrieval. We have developed scalable image representations and algorithms for both indexing and retrieval for content-based image retrieval. This technology has been transferred to Bing to enable indexing and searching billions of images by content, which has various applications. One feature released in Bing is "Find more sizes" where a user can find different sizes of near and partial duplicates of a given image.
Large-Scale Face Image Retrieval. Face images are popular on the web. We have developed customized face features and indexing/query pipeline to enable web-scale face image recognition and retrieval.
ImageGraph: Large-Scale Image Clustering. Our goal is to clustering the whole web images. As a first step, we have developed an efficient hashing scheme to discovering partial-duplicate web images.
- Zhong Wu, Qifa Ke, Jian Sun, and Heung-Yeung Shum, Scalable Face Image Retrieval with Identity-Based Quantization and Multireference Reranking, in IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), vol. 33, no. 10, pp. 1991 -2001, IEEE, October 2011
- David C. Lee, Qifa Ke, and Michael Isard, Partition Min-Hash for Partial Duplicate Image Discovery, in Proceedings of the European Conference on Computer Vision (ECCV 2010), Springer Verlag, September 2010
- Zhong Wu, Qifa Ke, Jian Sun, and Heung-Yeung Shum, Scalable Face Image Retrieval with Identity-Based Quantization and Multi-Reference Re-ranking, in CVPR 2010, IEEE Computer Society, June 2010
- Zhong Wu, Qifa Ke, Jian Sun, and Heung-Yeung Shum, A Multi-sample, Multi-tree Approach to Bag-of-words Image Representation for Image Retrieval, in The 12th International Conference on Computer Vision (ICCV), September 2009
- Zhong Wu, Qifa Ke, Michael Isard, and Jian Sun, Bundling Features for Large Scale Partial-DuplicateWeb Image Search, in CVPR 2009, IEEE, June 2009
Intern
Zhong Wu (2008,2009), David Lee (2009)
