期刊
SCIENCE CHINA-INFORMATION SCIENCES
卷 66, 期 11, 页码 -出版社
SCIENCE PRESS
DOI: 10.1007/s11432-022-3686-1
关键词
saliency detection; convolutional neural network; regional feature mapping; co-saliency detection; deep learning
This study proposes a novel end-to-end trainable network for co-saliency detection within a single image. The network combines bottom-up and top-down strategies by using ground-truth masks as top-down guidance and constructing triplet proposals for regional feature mapping and clustering.
Co-saliency detection within a single image is a common vision problem that has not yet been well addressed. Existing methods often used a bottom-up strategy to infer co-saliency in an image in which salient regions are firstly detected using visual primitives such as color and shape and then grouped and merged into a co-saliency map. However, co-saliency is intrinsically perceived complexly with bottom-up and top-down strategies combined in human vision. To address this problem, this study proposes a novel end-to-end trainable network comprising a backbone net and two branch nets. The backbone net uses ground-truth masks as top-down guidance for saliency prediction, whereas the two branch nets construct triplet proposals for regional feature mapping and clustering, which drives the network to be bottom-up sensitive to co-salient regions. We construct a new dataset of 2019 natural images with co-saliency in each image to evaluate the proposed method. Experimental results show that the proposed method achieves state-of-the-art accuracy with a running speed of 28 fps.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据