ETH-CVL @ MediaEval 2015: Learning Objective Functions for Improved Image Retrieval

In this paper, we present a method to select a rened subset of images, given an initial list of retrieved images. The goal of any image retrieval system is to present results that are maximally relevant as well as diverse. We formulate this as a subset selection problem and we address it using submodularity. In order to select the best subset, we learn an objective function as a linear combination of submodular functions. This objective quanties how relevant and representative a selected subset it. Using this method we obtain promising results at MediaEval 2015.

[1]  Andreas Krause,et al.  Submodular Function Maximization , 2014, Tractability.

[2]  Hui Lin,et al.  Learning Mixtures of Submodular Shells with Application to Document Summarization , 2012, UAI.

[3]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[4]  Pushmeet Kohli,et al.  Tractability: Practical Approaches to Hard Problems , 2013 .

[5]  Shih-Fu Chang,et al.  Image Retrieval: Current Techniques, Promising Directions, and Open Issues , 1999, J. Vis. Commun. Image Represent..

[6]  Trevor Darrell,et al.  DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition , 2013, ICML.

[7]  Yiannis Kompatsiaris,et al.  Improving Diversity in Image Search via Supervised Relevance Scoring , 2015, ICMR.

[8]  Jeff A. Bilmes,et al.  A Submodular-supermodular Procedure with Applications to Discriminative Structure Learning , 2005, UAI.

[9]  Yoram Singer,et al.  Adaptive Subgradient Methods for Online Learning and Stochastic Optimization , 2011, J. Mach. Learn. Res..

[10]  Yongdong Zhang,et al.  Multi-task deep visual-semantic embedding for video thumbnail selection , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[11]  Bogdan Ionescu,et al.  Retrieving Diverse Social Images at MediaEval 2017: Challenges, Dataset and Evaluation , 2017, MediaEval.

[12]  Luc Van Gool,et al.  Video summarization by learning submodular mixtures of objectives , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[13]  M. L. Fisher,et al.  An analysis of approximations for maximizing submodular set functions—I , 1978, Math. Program..

[14]  Nicu Sebe,et al.  Content-based multimedia information retrieval: State of the art and challenges , 2006, TOMCCAP.