首页|基于特征注意力提纯的显著性目标检测模型

基于特征注意力提纯的显著性目标检测模型

扫码查看
近年来,显著性目标检测技术取得了巨大进展,其中如何选择并有效集成多尺度特征扮演了重要角色.针对现有特征集成方法可能导致的信息冗余问题,提出了一种基于特征注意力提纯的显著性检测模型.首先,在解码器中采用一个全局特征注意力引导模块(GAGM)对带有语义信息的深层特征进行注意力机制处理,得到全局上下文信息;然后,通过全局引导流将其送入解码器各层进行监督训练;最后,利用多尺度特征融合模块(FAM)对编码器提取出的多尺度特征与全局上下文信息进行有效集成,并在网格状特征提纯模块(MFPM)中进行进一步细化,以生成清晰、完整的显著图.在5个公开数据集上进行实验,结果表明,所提模型优于现有的其他显著性检测方法,并且处理速度快,当处理320 × 320尺寸的图像时,能以30帧以上的速度运行.
Salient Object Detection Based on Feature Attention Purification
In recent years,salient object detection technology has made great progress,and how to select and effectively integrate multi-scale features plays an important role.Aiming at the information redundancy problem that may be caused by existing fea-ture integration methods,a saliency detection model based on feature attention refinement is proposed.First,in the decoder,a global feature attention guidance module(GAGM)is used to process the deep features with semantic information through the at-tention mechanism to obtain global context information,and then these information is sent to each layer of the decoder for super-vision through the global guidance flow train.The multi-scale features extracted by the encoder and the global context information are then effectively integrated using the multi-scale feature aggregation module(FAM),and further refined in the mesh feature purification module(MFPM)to generate clear and complete salient features.Experimental results on 5 public datasets demon-strate that the proposed model outperforms other existing saliency object detection methods.Besides,the processing speed of our approach is also very fast,it can run at a speed of more than 30 FPS when processing a 320 X 320 image.

Salient object detectionAttention mechanismMulti-scale feature fusionFeature selectionMesh feature purification

白雪飞、申悟呈、王文剑

展开 >

山西大学计算机与信息技术学院 太原 030006

计算智能与中文信息处理教育部重点实验室(山西大学)太原 030006

显著性目标检测 注意力机制 多尺度特征融合 特征选择 网格状特征提纯

国家自然科学基金国家自然科学基金国家自然科学基金国家自然科学基金山西省重点研发计划山西省回国留学人员科研项目

61703252U21A2051362076154622761612021021504010132022-008

2024

计算机科学
重庆西南信息有限公司(原科技部西南信息中心)

计算机科学

CSTPCD北大核心
影响因子:0.944
ISSN:1002-137X
年,卷(期):2024.51(5)
  • 44