Fine-Grained Classification via Hierarchical Feature Covariance Attention Module

Abstract

Fine-Grained Visual Classification (FGVC) has consistently been challenging in various domains, such as aviation and animal breeds. It is mainly due to the FGVC's criteria that differ with a considerably small range or subtle pattern differences. In the deep convolutional neural network, the covariance between feature maps positively affects the selection of features to learn discriminative regions automatically. In this study, we propose a method for a finegrained classification model by inserting an attention module that uses covariance characteristics. Specifically, we introduce a feature map attention module (FCA) to extract the feature map between convolution blocks, constituting the existing classification model. The FCA module then applies the corresponding value of the covariance matrix to the channel to focus on the salient area. We demonstrate the need for fine-grained classification in a hierarchical manner by focusing on the diverse scale representation. Additionally, we implemented two ablation studies to show how each suggested strategy affects classification performance. Our experiments are conducted on three datasets, CUB-200-2011, Stanford Cars, and FGVC-Aircraft, primarily used for fine-grained classification tasks. Our method outperforms the state-of-the-art models by a margin of 0.4%, 1.1%, and 1.4%.

Multistage Attention Region Supplement Transformer for Fine-grained Visual Categorization

“…The SR-GNN [39] model proposes a gate-controlled attention pool for automatic aggregation of relationship-aware features to capture subtle component features in the most relevant image regions, while the CAMF [40] module extracts complementary attention features while highlighting the most significant ones. To alleviate the computational resource demands of deep learning-based feature extraction methods, Jung YR et al [41] propose a low-cost feature extraction approach using the covariance matrix, achieving good performance in fine-grained visual classification.…”