TPAMI publishes article by XMU doctoral student
Recently, the IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), the top international academic journals in the field of computer science with an impact factor of 8.329, has published the latest research results of the research team headed by Prof. Ji Rongrong from Xiamen University. This article, entitled “Ordinal Constraint Binary Coding for Approximate Nearest Neighbor Search”, proposes a novel ranking-preserving hashing method, dubbed Ordinal Constraint Hashing (OCH), which efficiently learns the optimal hashing functions with a graph-based approximation to embed the ordinal relations. The core idea is to reduce the size of ordinal graph with ordinal constraint projection, which preserves the ordinal relations through a small data set (such as clusters or random samples). In particular, to learn such hash functions effectively, the research team further relaxes the discrete constraints and designs a specific stochastic gradient decent algorithm for optimization. The experimental results on three large-scale visual search benchmark datasets, i.e. LabelMe, Tiny100K and GIST1M, showed that the proposed OCH method can achieve superior performance over the state-of-the-arts approaches. Currently some of the code of the article has been released as open source and previous research results had been published on CCF-A journals such as CVPR/AAAI/IJCAI/TIP.
This article is the result of the collaborative effort of Liu Hong, a doctoral student from Xiamen University, his tutor Prof. Ji Rongrong (Corresponding author), Wang Jingdong, a researcher from Microsoft Research Asia and Shen Chunhua, a professor of Adelaide University. It is the first time for a graduate student from XMU to publish as the first author in the top journal in the field of computers, which marks a milestone in the development of XMU information science with regard to its graduate training program. The research is funded by the National Natural Science Fund(No.U1705262，No.61772443，and No.61572410), the National key research and development project (No.2017YFC0113000, and No.2016YFB1001503) etc.