期刊
IEEE TRANSACTIONS ON IMAGE PROCESSING
卷 29, 期 -, 页码 8407-8416出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIP.2020.3014734
关键词
Image reconstruction; Feature extraction; Object detection; Topology; Image color analysis; Machine learning; Diversity reception; Disentangle; RGBD; saliency detection
资金
- Research Grants Council of Hong Kong [CityU 11203619]
- Delta-NTU Corporate Lab for Cyber-Physical Systems
- National Research Foundation (NRF) Singapore under the Corp Lab@University Scheme
- National Research Foundation Singapore under its AI Singapore Program [AISG-RP-2018-003]
- MOE Tier-1 Research [RG22/19 (S)]
- Delta Electronics Inc.
Depth is beneficial for salient object detection (SOD) for its additional saliency cues. Existing RGBD SOD methods focus on tailoring complicated cross-modal fusion topologies, which although achieve encouraging performance, are with a high risk of over-fitting and ambiguous in studying cross-modal complementarity. Different from these conventional approaches combining cross-modal features entirely without differentiating, we concentrate our attention on decoupling the diverse cross-modal complements to simplify the fusion process and enhance the fusion sufficiency. We argue that if cross-modal heterogeneous representations can be disentangled explicitly, the cross-modal fusion process can hold less uncertainty, while enjoying better adaptability. To this end, we design a disentangled cross-modal fusion network to expose structural and content representations from both modalities by cross-modal reconstruction. For different scenes, the disentangled representations allow the fusion module to easily identify and incorporate desired complements for informative multi-modal fusion. Extensive experiments show the effectiveness of our designs and a large outperformance over state-of-the-art methods.
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