当前位置: X-MOL 学术J. Visual Commun. Image Represent. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Attention-based contextual interaction asymmetric network for RGB-D saliency prediction
Journal of Visual Communication and Image Representation ( IF 2.6 ) Pub Date : 2020-12-05 , DOI: 10.1016/j.jvcir.2020.102997
Xinyue Zhang , Ting Jin , Wujie Zhou , Jingsheng Lei

Saliency prediction on RGB-D images is an underexplored and challenging task in computer vision. We propose a channel-wise attention and contextual interaction asymmetric network for RGB-D saliency prediction. In the proposed network, a common feature extractor provides cross-modal complementarity between the RGB image and corresponding depth map. In addition, we introduce a four-stream feature-interaction module that fully leverages multiscale and cross-modal features for extracting contextual information. Moreover, we propose a channel-wise attention module to highlight the feature representation of salient regions. Finally, we refine coarse maps through a corresponding refinement block. Experimental results show that the proposed network achieves a performance comparable with state-of-the-art saliency prediction methods on two representative datasets.



中文翻译:

基于注意力的上下文交互非对称网络用于RGB-D显着性预测

在计算机视觉中,对RGB-D图像进行显着性预测是一项尚未开发且具有挑战性的任务。我们提出了针对RGB-D显着性预测的逐通道注意和上下文交互不对称网络。在所提出的网络中,公共特征提取器在RGB图像和相应的深度图之间提供了交叉模式的互补性。此外,我们引入了一个四流功能交互模块,该模块充分利用多尺度和跨模式功能来提取上下文信息。此外,我们提出了一个通道注意模块来突出显示显着区域的特征。最后,我们通过相应的细化块细化粗图。

更新日期:2020-12-09
down
wechat
bug