Associating Multi-Scale Receptive Fields for Fine-grained Recognition

TL;DR

This paper introduces a cross-layer non-local module that models interactions between multi-scale features, significantly improving fine-grained image recognition by capturing spatial dependencies across different layers.

Contribution

We propose a novel cross-layer non-local module that effectively associates multi-scale receptive fields for enhanced fine-grained recognition.

Findings

Achieves state-of-the-art results on three benchmark datasets.

Reduces aggregation cost by using low-dimensional deep layers.

Builds spatial dependencies among multi-level features.

Abstract

Extracting and fusing part features have become the key of fined-grained image recognition. Recently, Non-local (NL) module has shown excellent improvement in image recognition. However, it lacks the mechanism to model the interactions between multi-scale part features, which is vital for fine-grained recognition. In this paper, we propose a novel cross-layer non-local (CNL) module to associate multi-scale receptive fields by two operations. First, CNL computes correlations between features of a query layer and all response layers. Second, all response features are weighted according to the correlations and are added to the query features. Due to the interactions of cross-layer features, our model builds spatial dependencies among multi-level layers and learns more discriminative features. In addition, we can reduce the aggregation cost if we set low-dimensional deep layer as query layer. Experiments are conducted to show our model achieves or surpasses state-of-the-art results on three benchmark datasets of fine-grained classification. Our codes can be found at github.com/FouriYe/CNL-ICIP2020.