Pyramid attention object detection network with multi-scale feature fusion. (December 2022)

• Record Type: Journal Article
• Authors: Chen, Xiu; Li, Yujie; Nakatoh, Yoshihisa
• Is Part Of: Computers & electrical engineering. Volume 104:Part A(2022)
• Publication Date: 2022-12
• Subjects: Multi-scale features -- Small objects -- Object detection -- Contextual information -- Feature pyramid -- Attention module; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2022.108436
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: A multi-scale feature fusion pyramid attention module is proposed to better capture global and local features and improve the performance of object detection. Shows good results on PASCAL VOC and COCO dataset object detection tasks. By adding the weight calculation of the attention mechanism to the residual block, a more delicate feature representation is obtained. Abstract: With the development of deep learning, object detection has made substantial progress. However, when the object to be detected in the image is small or partially occluded, the detection network often fails to detect it successfully. We propose a multi-scale feature fusion pyramid attention module, which effectively combines the global average pooling results of multiple scales with the upper features in the residual blocks of the feature extraction network to obtain more spatial context information in the original feature map. We added the multi-scale feature fusion pyramid attention module proposed in this paper based on YoloV3 and conducted experiments on the PASCALL VOC and MS COCO datasets. The experimental results show that the attention module can effectively help the network detect small objects and accurately detect partially occlusion objects. Graphical abstract: Image, graphical abstract