Infomax principle based pooling of deep convolutional activations for image retrieval
RIS ID
116794
Abstract
Neural activations produced by deep convolutional networks have recently become state-of-the-art representation for image retrieval. To obtain a global image representation, sum-pooling has been frequently used to aggregate activations of convolutional feature maps. This work first presents an understanding on the effectiveness of sum-pooling via probabilistic interpretation, by proving that sum-pooling is an upper bound of the probability that a visual pattern is present in an image. To further answer the optimality of sum-pooling, a quantitative analysis based on the Infomax principle in neural networks is provided. It shows that sum-pooling aligns well with the leading eigenvector of principal component analysis (PCA) applied to the activations of a feature map. Moreover, considering the 2D matrix structure of feature maps, a two-directional 2DPCA-based pooling scheme is proposed to aggregate the convolutional activations. Experiments on multiple benchmark image retrieval datasets demonstrate the above analysis and the superiority of the proposed pooling scheme.
Publication Details
Gao, Z., Wang, L., Zhou, L. & Yang, M. (2017). Infomax principle based pooling of deep convolutional activations for image retrieval. IEEE International Conference on Multimedia and Expo (ICME) (pp. 457-462). United States: IEEE.