分层双线性池化图像行为识别方法  

Hierarchical bilinear pooling method for image-based action recognition

作　　者：吴伟[1] 于嘉乐 Wu Wei;Yu Jiale(School of Computer,Inner Mongolia University,Hohhot 010021,China)

机构地区：[1]内蒙古大学计算机学院,呼和浩特010021

出　　处：《电子测量与仪器学报》2021年第3期152-157,共6页Journal of Electronic Measurement and Instrumentation

基　　金：国家自然科学基金(61763035);内蒙古自然科学基金(2020MS06006)资助项目。

摘　　要：基于图像的行为识别由于受到类内图像背景信息的差异性和类间行为的相似性影响,至今仍然是一项极具挑战性的任务。某些行为类别在人物姿态、表情动作方面十分相似,因此对图像中各种富含语义信息的部位提取显著性特征对于提高行为识别的精度至关重要。借鉴双线性池化模型在细粒度分类中的优势,同时为避免该模型包含大量背景噪声而影响识别精度,提出一种改进的双线性池化模型用于图像行为识别。该模型利用通道和空间注意力机制关注图像中的重要目标,并通过集成多层注意力掩码图来生成RoI,这可以有效抑制图像中的背景噪声信息,提高行为识别的准确性。最终提出的方法在Stanford-40 dataset获得了85.24%准确率,同时在自定义的60类行为数据集上获得了84.57%的准确率。Image-based action recognition is still a very challenging task because it is disturbed by the differences in the background information of the images in the class and the similarity of the behavior between the classes. Some action categories are very similar in terms of human poses and facial expressions, so extracting salient features from various parts of the image that are rich in semantic information is essential to improve the accuracy of action recognition. Drawing on the advantages of the bilinear pooling model in fine-grained image classification, and to avoid this model which containing a lot of background noise to affect the recognition accuracy, an improved bilinear pooling model is proposed for action recognition in the paper. The model uses channel and spatial-wise attention mechanism to focus on the important targets in the image, and generates RoI by integrating multi-layer attention mask, which can effectively suppress the background noise information in the image and improve the accuracy of action recognition. Our method achieves the accuracy of 85.24% on the Stanford-40 dataset, and the accuracy of 84.57% on the custom 60 kind of action dataset.

关 键 词：图像行为识别 分层双线性池化 注意力机制 掩码聚合 

分 类 号：TP37[自动化与计算机技术—计算机系统结构]