Fisher Kernel for Deep Neural Activations

TL;DR

This paper introduces a method combining multi-scale CNN activations with Fisher kernel aggregation, significantly improving image recognition performance by leveraging both local detail and mid-level features.

Contribution

It proposes an efficient multi-scale dense activation extraction and a modified Fisher kernel for CNN features, enhancing image representation for recognition tasks.

Findings

+17.76% accuracy on MIT Indoor 67

+7.18% mAP on PASCAL VOC 2007

Significant performance improvements in visual recognition

Abstract

Compared to image representation based on low-level local descriptors, deep neural activations of Convolutional Neural Networks (CNNs) are richer in mid-level representation, but poorer in geometric invariance properties. In this paper, we present a straightforward framework for better image representation by combining the two approaches. To take advantages of both representations, we propose an efficient method to extract a fair amount of multi-scale dense local activations from a pre-trained CNN. We then aggregate the activations by Fisher kernel framework, which has been modified with a simple scale-wise normalization essential to make it suitable for CNN activations. Replacing the direct use of a single activation vector with our representation demonstrates significant performance improvements: +17.76 (Acc.) on MIT Indoor 67 and +7.18 (mAP) on PASCAL VOC 2007. The results suggest that our proposal can be used as a primary image representation for better performances in visual recognition tasks.