A fast X-shaped foreground segmentation network with CompactASPP. (January 2021)

• Record Type: Journal Article
• Authors: Zhang, Jin; Wang, Shuaihui; Qiu, Junyang; Pan, Xinran; Zou, Junhua; Duan, Yexin; Pan, Zhisong; Li, Yang
• Is Part Of: Engineering applications of artificial intelligence. Volume 97(2021)
• Publication Date: 2021-01
• Subjects: 00-01 -- 99-00; Foreground segmentation -- ASPP -- Multi-scale feature representation -- Atrous convolution; Engineering -- Data processing -- Periodicals; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Ingénierie -- Informatique -- Périodiques; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; Artificial intelligence; Engineering -- Data processing; Expert systems (Computer science); Periodicals; 620.00285
• DOI: 10.1016/j.engappai.2020.104077
• Languages: English
• ISSNs: 0952-1976

Abstract

Abstract: Foreground segmentation models are designed to extract moving objects of varying sizes from the background, which can benefit from representations of various scales. As an effective module for capturing multi-scale contexts, Atrous Spatial Pyramid Pooling (ASPP) convolves a final feature representation via multiple parallel atrous convolutions with different dilation rates. However, as the dilation rate increases, ASPP gradually loses its large-scale modeling ability because the sampling of atrous kernel becomes progressively sparse within the receptive field. To solve this problem, we design a CompactASPP module to convolve feature maps compactly . Without significantly increasing the module size, the CompactASPP can encode multi-scale features from all neurons within the receptive field rather than from neurons in several sparsely distributed positions. Furthermore, we leverage CompactASPP modules to enhance our previous X-Net. The proposed Fast X-Net substantially improves the segmentation speed by over 63.6% and attains new state-of-the-art performances on CDnet2014, SBI2015 and UCSD benchmarks.