Efficient Codebook and Factorization for Second Order Representation Learning
This work provides a solution for improving representation compactness in computer vision applications, though it is incremental as it builds on existing second-order and factorization approaches.
The paper tackles the problem of learning compact yet rich representations for tasks like image retrieval by addressing the high dimensionality of second-order methods. It introduces a method combining codebook and factorization strategies, achieving state-of-the-art results on three image retrieval datasets while maintaining performance with few additional parameters.
Learning rich and compact representations is an open topic in many fields such as object recognition or image retrieval. Deep neural networks have made a major breakthrough during the last few years for these tasks but their representations are not necessary as rich as needed nor as compact as expected. To build richer representations, high order statistics have been exploited and have shown excellent performances, but they produce higher dimensional features. While this drawback has been partially addressed with factorization schemes, the original compactness of first order models has never been retrieved, or at the cost of a strong performance decrease. Our method, by jointly integrating codebook strategy to factorization scheme, is able to produce compact representations while keeping the second order performances with few additional parameters. This formulation leads to state-of-the-art results on three image retrieval datasets.