arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2108.01971cs.CV
收藏

基于跨模态差异交互网络的RGB-D显著目标检测

Cross-modality Discrepant Interaction Network for RGB-D Salient Object Detection

Beijing Jiaotong University(北京交通大学);Meituan(美团);City University of Hong Kong(香港城市大学)

Chen Zhang, Runmin Cong, Qinwei Lin, Lin Ma, Feng Li, Yao Zhao, Sam Kwong

摘要:

深度图的普及和推广为显著目标检测(SOD)带来了新的生机和活力,大量的RGB-D SOD算法被提出,主要集中在如何更好地融合RGB图像和深度图的跨模态特征。对于特征编码器中的跨模态交互,现有方法要么不加区别地对待RGB和深度模态,要么习惯性地将深度线索作为RGB分支的辅助信息。与之不同的是,我们重新考虑了两种模态的状态,并提出了一种新的跨模态差异交互网络(CDINet)的RGB-D SOD,根据不同层的特征表示差异模型的依赖性的两种模态。为此,设计了两个组件来实现有效的跨模态交互:1)RGB诱导的细节增强(RDE)模块利用RGB模态在低级编码器阶段增强深度特征的细节。2)深度诱导语义增强(DSE)模块将深度特征的对象定位和内部一致性传递到高级编码器阶段的RGB分支。此外,我们还设计了一种密集解码重建(DDR)结构,它通过结合多层次的编码器特征来构建一个语义块,以提升特征解码中的跳过连接。在五个基准数据集上进行的大量实验表明,我们的网络在定量和定性方面都优于$15$最先进的方法。我们的代码可在https://rmcong.github.io/proj_CDINet.html上公开获取。

英文摘要:

The popularity and promotion of depth maps have brought new vigor and vitality into salient object detection (SOD), and a mass of RGB-D SOD algorithms have been proposed, mainly concentrating on how to better integrate cross-modality features from RGB image and depth map. For the cross-modality interaction in feature encoder, existing methods either indiscriminately treat RGB and depth modalities, or only habitually utilize depth cues as auxiliary information of the RGB branch. Different from them, we reconsider the status of two modalities and propose a novel Cross-modality Discrepant Interaction Network (CDINet) for RGB-D SOD, which differentially models the dependence of two modalities according to the feature representations of different layers. To this end, two components are designed to implement the effective cross-modality interaction: 1) the RGB-induced Detail Enhancement (RDE) module leverages RGB modality to enhance the details of the depth features in low-level encoder stage. 2) the Depth-induced Semantic Enhancement (DSE) module transfers the object positioning and internal consistency of depth features to the RGB branch in high-level encoder stage. Furthermore, we also design a Dense Decoding Reconstruction (DDR) structure, which constructs a semantic block by combining multi-level encoder features to upgrade the skip connection in the feature decoding. Extensive experiments on five benchmark datasets demonstrate that our network outperforms $15$ state-of-the-art methods both quantitatively and qualitatively. Our code is publicly available at: https://rmcong.github.io/proj_CDINet.html.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑