arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.01779cs.CV
收藏

基于深度质量的有效RGB-D显著目标检测特征处理

Depth Quality-Inspired Feature Manipulation for Efficient RGB-D Salient Object Detection

Sichuan University(四川大学);Inception Institute of AI(Inception人工智能研究院)

Wenbo Zhang, Ge-Peng Ji, Zhuo Wang, Keren Fu, Qijun Zhao

摘要:

RGB-D显著性目标检测(SOD)由于具有额外的深度信息而使传统的RGB SOD受益,近年来吸引了越来越多的研究兴趣。然而,现有的RGB-D SOD模型在效率和准确性方面往往表现不佳,这阻碍了它们在移动设备和现实问题上的潜在应用。一个潜在的挑战是,当模型被简化为只有很少的参数时,模型的准确性通常会降低。为了解决这个难题,也受到深度质量是影响准确性的关键因素的启发,我们提出了一种新的深度质量启发的特征操作(DQFM)过程,它本身是高效的,可以作为过滤深度特征的门控机制,以大大提高准确性。DQFM采用低级别RGB和深度特征的对齐,以及对深度流的整体关注来显式控制和增强跨模态融合。我们嵌入DQFM以获得一个高效的轻量级模型,称为DFM-Net,其中我们还设计了一个定制的深度骨干和一个两级解码器,以进一步提高效率。大量的实验结果表明,我们的DFM-Net达到了最先进的精度相比,现有的非有效的模型,同时运行在140毫秒的CPU(2.2 $\times$比以前最快的有效模型),只有$\sim$ 8.5 Mb的模型大小(14.9%的以前最轻)。我们的代码将在https://github.com/zwbx/DFM-Net上提供。

英文摘要:

RGB-D salient object detection (SOD) recently has attracted increasing research interest by benefiting conventional RGB SOD with extra depth information. However, existing RGB-D SOD models often fail to perform well in terms of both efficiency and accuracy, which hinders their potential applications on mobile devices and real-world problems. An underlying challenge is that the model accuracy usually degrades when the model is simplified to have few parameters. To tackle this dilemma and also inspired by the fact that depth quality is a key factor influencing the accuracy, we propose a novel depth quality-inspired feature manipulation (DQFM) process, which is efficient itself and can serve as a gating mechanism for filtering depth features to greatly boost the accuracy. DQFM resorts to the alignment of low-level RGB and depth features, as well as holistic attention of the depth stream to explicitly control and enhance cross-modal fusion. We embed DQFM to obtain an efficient light-weight model called DFM-Net, where we also design a tailored depth backbone and a two-stage decoder for further efficiency consideration. Extensive experimental results demonstrate that our DFM-Net achieves state-of-the-art accuracy when comparing to existing non-efficient models, and meanwhile runs at 140ms on CPU (2.2$\times$ faster than the prior fastest efficient model) with only $\sim$8.5Mb model size (14.9% of the prior lightest). Our code will be available at https://github.com/zwbx/DFM-Net.

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。

↑