Article
Computer Science, Artificial Intelligence
Jia Li, Shengye Qiao, Zhirui Zhao, Chenxi Xie, Xiaowu Chen, Changqun Xia
Summary: This article introduces a lightweight framework for salient object detection, which addresses the dilution of semantic context, loss of spatial structure, and absence of boundary detail by decoupling the U-shape structure into three branches. The proposed Scale-Adaptive Pooling Module is used to obtain multi-scale receptive field. Experimental results demonstrate that the method achieves a better balance between efficiency and accuracy.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Computer Science, Information Systems
Xiaofang Li, Yi Wang, Tianzhu Wang, Ruili Wang
Summary: In this work, an effective and flexible spatial frequency enhancement (SFE) module based on generalized Oct-convolution is proposed. It can extract and incorporate multiple spatial frequency information from different feature maps and output comprehensive and compact frequency features. A spatial frequency enhanced network (SFENet) is then designed, which adopts two SFE modules to refine high- and low-frequency salient features and integrates them into a final full-band saliency prediction.
INFORMATION SCIENCES
(2023)
Article
Computer Science, Artificial Intelligence
Zhengzheng Tu, Zhun Li, Chenglong Li, Jin Tang
Summary: In this study, we propose a novel deep correlation network for RGBT Salient Object Detection (SOD). The network explores the correlations between RGB and thermal modalities, and incorporates a modality alignment module and a bi-directional decoder model to handle unaligned image pairs and enhance feature representation. Experimental results show that our method outperforms state-of-the-art methods on three benchmark datasets.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Computer Science, Information Systems
Shuaixiong Hui, Qiang Guo, Xiaoyu Geng, Caiming Zhang
Summary: Feature refinement and fusion are crucial steps in SOD. This article proposes MGuid-Net, a novel multi-guidance SOD model that utilizes multiple guidance mechanisms. It incorporates edge features alongside saliency features and includes self-guidance and cross-guidance modules to refine and fuse the features. The model also incorporates an accumulative guidance module and a pixelwise contrast loss function to better integrate and retain details. Experimental results show that the proposed model outperforms state-of-the-art models on benchmark datasets.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Zhaojian Yao, Luping Wang
Summary: This research focuses on the use of boundary information in saliency detection and proposes a new network structure to generate more accurate saliency maps by learning boundary features. Experimental results demonstrate that the proposed method outperforms 15 state-of-the-art methods on benchmark datasets.
IEEE TRANSACTIONS ON MULTIMEDIA
(2022)
Article
Computer Science, Artificial Intelligence
Xiaowei Chen, Qing Zhang, Liqian Zhang
Summary: The proposed edge-aware salient object detection network utilizes high-level semantic information to assist feature selection and locates salient objects by extracting multi-scale features and emphasizing important feature channels. It adopts a context guidance strategy to fuse high-level and low-level information and supervises the generation of low-level edge information.
IMAGE AND VISION COMPUTING
(2021)
Article
Computer Science, Artificial Intelligence
Xian Fang, Jinchao Zhu, Xiuli Shao, Hongpeng Wang
Summary: In this paper, we propose a novel network model LC(3)Net, equipped with the components of FCB, DCM, and BCD, to address the issues in utilizing contextual information. Extensive experiments demonstrate the superior performance of our method compared to 20 state-of-the-art methods.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Hardware & Architecture
Botong Zhang, Lihua Tian, Chen Li, Yi Yang
Summary: In this paper, a novel network called HENet is proposed to achieve better prediction results by extracting and utilizing features of different layers. The feature extraction module and multi-layer feature supplementary module are used to obtain location and detailed information. Furthermore, the trisection dilated convolution module is proposed to expand the receptive field of features. Experimental results demonstrate the superiority of our method on 4 datasets.
COMPUTERS & ELECTRICAL ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Junbin Yuan, Lifang Xiao, Kanoksak Wattanachote, Qingzhen Xu, Xiaonan Luo, Yongyi Gong
Summary: In this paper, a fixation guidance network (FGNet) is proposed for salient object detection, which utilizes fixation prediction to guide both salient object detection and edge detection. The network consists of a multi-branch structure for multi-task detection, a fixation guidance module to guide detection, and a multi-resolution feature interaction module for optimizing the representations. Experimental results show that the proposed method outperforms existing algorithms.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Automation & Control Systems
Jie Wang, Kechen Song, Yanqi Bao, Yunhui Yan, Yahong Han
Summary: This paper introduces a unidirectional RGB-T salient object detection network with intertwined driving of encoding and fusion. By using transformer as the network backbone, it solves the problem of CNNs' difficulty in establishing long-range dependencies. Furthermore, by constructing a unidirectional architecture and using local detail-driven modules, it improves the drawbacks of the encoder-decoder architecture and enhances the performance of the network.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Jingyu Wu, Fuming Sun, Rui Xu, Jie Meng, Fasheng Wang
Summary: This paper proposes a strategy of aggregation and interaction to extract edge features, depth features, and salient features while maintaining local details and fully extracting global information. By extracting and fusing features in the learning process, it addresses the issues of multi-scale problem and information redundancy, achieving excellent performance in salient object detection.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Fushuo Huo, Xuegui Zhu, Bingheng Li
Summary: This paper proposes the Three-stream Interaction Decoder Network (TIDNet) for the RGB-T SOD task. By utilizing a three-stream interaction decoder in the encoder branches, we are able to explore saliency in depth and capture salient cues from both single and multi-modalities. Our method outperforms state-of-the-art methods in comprehensive experiments.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Engineering, Civil
Ning Jia, Yougang Sun, Xianhui Liu
Summary: This paper proposes a traffic salient object detection method that can detect complete objects that attract human attention in natural traffic scenes, providing assistance for target recognition tasks in the domain of intelligent driving.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Yang Yang, Qi Qin, Yongjiang Luo, Yi Liu, Qiang Zhang, Jungong Han
Summary: This paper presents a Bi-directional Progressive Guidance Network (BPGNet) for RGB-D salient object detection, which involves the qualities of RGB and depth images. The network employs a bi-directional framework based on progressive guidance strategy to extract and enhance unimodal features, in order to address the impact of RGB image quality on saliency detection.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Chang Xu, Qingwu Li, Qingkai Zhou, Xiongbiao Jiang, Dabing Yu, Yaqin Zhou
Summary: RGB-thermal salient object detection has unique advantages in handling challenging scenes, but existing methods often overlook the differences between imaging mechanisms and thermal image characteristics, resulting in unsatisfactory performance. To address this, an asymmetric cross-modal activation network is proposed to achieve more effective RGB-T SOD by exploiting the interactions of modality-specific features.
KNOWLEDGE-BASED SYSTEMS
(2022)