BA^2M: A Batch Aware Attention Module for Image Classification

03/28/2021
by   Qishang Cheng, et al.
0

The attention mechanisms have been employed in Convolutional Neural Network (CNN) to enhance the feature representation. However, existing attention mechanisms only concentrate on refining the features inside each sample and neglect the discrimination between different samples. In this paper, we propose a batch aware attention module (BA2M) for feature enrichment from a distinctive perspective. More specifically, we first get the sample-wise attention representation (SAR) by fusing the channel, local spatial and global spatial attention maps within each sample. Then, we feed the SARs of the whole batch to a normalization function to get the weights for each sample. The weights serve to distinguish the features' importance between samples in a training batch with different complexity of content. The BA2M could be embedded into different parts of CNN and optimized with the network in an end-to-end manner. The design of BA2M is lightweight with few extra parameters and calculations. We validate BA2M through extensive experiments on CIFAR-100 and ImageNet-1K for the image recognition task. The results show that BA2M can boost the performance of various network architectures and outperforms many classical attention methods. Besides, BA2M exceeds traditional methods of re-weighting samples based on the loss value.

READ FULL TEXT

page 1

page 2

page 3

research
11/14/2022

PKCAM: Previous Knowledge Channel Attention Module

Recently, attention mechanisms have been explored with ConvNets, both ac...
research
05/23/2019

Spatial Group-wise Enhance: Improving Semantic Feature Learning in Convolutional Networks

The Convolutional Neural Networks (CNNs) generate the feature representa...
research
09/12/2023

IBAFormer: Intra-batch Attention Transformer for Domain Generalized Semantic Segmentation

Domain generalized semantic segmentation (DGSS) is a critical yet challe...
research
02/14/2022

Convolutional Neural Network with Convolutional Block Attention Module for Finger Vein Recognition

Convolutional neural networks have become a popular research in the fiel...
research
07/13/2020

Deep Reinforced Attention Learning for Quality-Aware Visual Recognition

In this paper, we build upon the weakly-supervised generation mechanism ...
research
12/12/2017

Direction-aware Spatial Context Features for Shadow Detection

Shadow detection is a fundamental and challenging task, since it require...
research
10/09/2020

Permuted AdaIN: Enhancing the Representation of Local Cues in Image Classifiers

Recent work has shown that convolutional neural network classifiers over...

Please sign up or login with your details

Forgot password? Click here to reset