Feature Reintegration over Differential Treatment: A Top-down and Adaptive Fusion Network for RGB-D Salient Object Detection
Feature Reintegration over Differential Treatment: A Top-down and Adaptive Fusion Network for RGB-D Salient Object Detection
复制标题
DOI：
10.1145/3394171.3413969
复制
发表时间：
2020-10
期刊：
影响因子：
--
通讯作者：
Miao Zhang;Yu Zhang;Yongri Piao;Beiqi Hu;Huchuan Lu
中科院分区：
文献类型：
--
作者：
Miao Zhang;Yu Zhang;Yongri Piao;Beiqi Hu;Huchuan Lu
大多数RGB-D显著对象检测（SOD）方法都采用相同的融合策略来探索每个层次上的跨模态互补信息。然而，这可能会忽略两种模态在不同水平上对预测的不同特征贡献。在本文中，我们提出了一种新的自顶向下的多层次融合结构，利用不同的融合策略，有效地探索低层次和高层次的功能。这是通过设计交织融合模块（IFM），以有效地整合全局信息和设计门控选择融合模块（GSFM），通过过滤掉不必要的RGB和深度数据有区别地选择有用的本地信息。此外，我们提出了一个自适应融合模块（AFM）重新整合融合的跨模态特征的每一个级别，以预测更准确的结果。在7个具有挑战性的基准数据集上进行的综合实验表明，我们的方法在14种最先进的RGB-D替代方法上具有竞争力的性能。
Most methods for RGB-D salient object detection (SOD) utilize the same fusion strategy to explore the cross-modal complementary information at each level. However, this may ignore different feature contributions from two modalities on different levels towards prediction. In this paper, we propose a novel top-down multi-level fusion structure where different fusion strategies are utilized to effectively explore the low-level and high-level features. This is achieved by designing the interweave fusion module (IFM) to effectively integrate the global information and designing the gated select fusion module (GSFM) to discriminatively select useful local information by filtering out the unnecessary one from RGB and depth data. Moreover, we propose an adaptive fusion module (AFM) to reintegrate the fused cross-modal features of each level to predict a more accurate result. Comprehensive experiments on 7 challenging benchmark datasets demonstrate that our method achieves the competitive performance over 14 state-of-the-art RGB-D alternative methods.