相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Automation & Control Systems
Bo Lyu et al.
Summary: This article introduces an automatic exploration method for efficient portrait parsing models, which can be easily deployed in edge computing or terminal devices. By using multiobjective reinforcement learning and neural architecture search, a balance between resource cost and performance is achieved, resulting in a series of excellent portrait parsing models.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Engineering, Electrical & Electronic
Gang Chen et al.
Summary: This paper proposes a novel Modality-Induced Transfer-Fusion Network (MITF-Net) for RGB-D and RGB-T salient object detection. The network is designed to capture the complementary information of multi-modality data and improve performance through modality transfer fusion and cycle-separated attention modules. The proposed MITF-Net achieves competitive and excellent performance on RGB-D and RGB-T salient object detection datasets.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2023)
Article
Automation & Control Systems
Kechen Song et al.
Summary: Visual perception is crucial for industrial information field, specifically in robotic grasping application. To achieve fast and accurate object detection for grasping, salient object detection (SOD) is employed. However, existing SOD methods still have limitations in practical application due to complex interference. To address this, a novel triple-modal images fusion strategy called visible-depth-thermal (VDT) SOD is proposed. Experimental results demonstrate that our method outperforms state-of-the-art approaches.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2023)
Article
Automation & Control Systems
Han Wang et al.
Summary: RGB-T salient object detection has achieved rapid development and excellent results in recent years. However, the current RGB-T datasets lack low-illumination data, leading to poor performance in detecting salient objects in extremely low-illumination scenes. To address this issue, we propose a T-aware guided early fusion network that leverages thermal images to enhance the detection performance of low-illumination data.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Bo Lyu et al.
Summary: This paper proposes the deployment of spectral graph convolutional networks on memristive crossbars and accelerates the computation using Sparse Laplace Matrix Reordering and Diagonal Block Matrix Multiplication. The simulated experiments achieve high overall accuracy and a reduction in memristor number.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Computer Science, Information Systems
Runmin Cong et al.
Summary: This paper explores the RGB-T salient object detection task and proposes a network model named TNet, which tackles the problem by introducing a global illumination estimation module and cross-modality interaction mechanisms. Experimental results demonstrate that TNet achieves competitive performance compared to other state-of-the-art methods.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Computer Science, Information Systems
Zhengzheng Tu et al.
Summary: Salient object detection in complex scenes is challenging, with RGBT SOD emerging as a new research direction. The VT5000 dataset contains 5000 spatially aligned RGBT image pairs and a baseline approach that utilizes multi-level feature extraction and attention mechanism for accurate RGBT SOD.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Computer Science, Artificial Intelligence
Nianchang Huang et al.
Summary: A novel end-to-end RGB-D salient object detection model is proposed in this paper, which combines a Semantic-Guided Modality-Weight Map Generation sub-network and a Bi-directional Multi-scale Cross-modal Feature Fusion module to effectively address the issue of poor input image quality affecting discriminative ability.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Aojun Gong et al.
Summary: This paper proposes an unsupervised method for RGB-T image saliency detection by constructing a graph model and fusing algorithms such as classification distance and sparse constraint to enhance detection accuracy.
APPLIED INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Jingyu Wu et al.
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
Engineering, Electrical & Electronic
Jie Wang et al.
Summary: This paper proposes a Cross-Guided Fusion Network (CGFNet) for RGB-T salient object detection, which combines RGB and thermal infrared (T) images to improve the detection performance of salient objects by leveraging the unique characteristics of different modalities.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Wujie Zhou et al.
Summary: ECFFNet is a RGB-T feature fusion network that achieves accurate detection of salient objects by combining thermal images and RGB images. It outperforms state-of-the-art methods by implementing cross-modality fusion, bilateral reversal fusion, and multilevel consistent fusion.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Wei Gao et al.
Summary: The use of complementary information, such as depth or thermal information, has shown benefits in salient object detection. However, the current approaches for RGB-D or RGB-T salient object detection problems are solved independently and they directly extract and fuse raw features. This work proposes a unified end-to-end framework that can simultaneously analyze RCB-D and RGB-T salient object detection tasks, and it introduces a multi-stage and multi-scale fusion network to effectively handle multi-modal features.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Fushuo Huo et al.
Summary: The study proposed an efficient encoder-decoder network named Context-guided Stacked Refinement Network (CSRNet), which reduces computational cost using a lightweight backbone and efficient decoder parts, fuses RGB and T modalities through the Context-guided Cross Modality Fusion (CCMF) module, and refines features progressively via the Stacked Refinement Network (SRN).
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Engineering, Electrical & Electronic
Liming Huang et al.
Summary: RGB and thermal infrared (RGBT) image saliency detection is a new direction in computer vision. This paper proposes an unsupervised RGBT saliency detection method based on multi-graph fusion and learning, which improves detection performance by fusing and learning from multi-modal images, and the effectiveness of the method is validated in experiments.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Lei Yang et al.
Summary: Defect detection is crucial for product quality control and repair decision-making. Nondestructive testing (NDT) is effective, but faces challenges such as complex backgrounds and class imbalance. Deep learning has improved automatic defect detection, but limitations remain due to insufficient processing of local contextual features. A novel nondestructive defect detection network, NDD-Net, incorporating an attention fusion block (AFB) and a residual dense connection convolution block (RDCCB), outperforms other related models in segmenting microdefects.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Yanhua Liang et al.
Summary: In this paper, a novel end-to-end unified framework is proposed for RGB-D and RGB-T salient object detection. The framework consists of three key components: multi-modal interactive attention unit, joint attention guided cross-modal decoding module, and multi-level feature progressive decoding module. Experimental results show that the proposed method outperforms other state-of-the-art methods and has good generalization.
Article
Engineering, Electrical & Electronic
Xiaofei Zhou et al.
Summary: In this paper, a novel feature aggregation network FANet is proposed for RGBD saliency detection, which includes a feature extraction module and an aggregation module. The FANet is able to adequately mine and fuse the geometric information and the appearance information contained in depth maps and RGB images, achieving compelling performance against the current state-of-the-art RGBD saliency models.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2022)
Article
Computer Science, Artificial Intelligence
Jiadong Chen et al.
Summary: This article introduces a new approach for designing memristor-based fully convolutional neural networks, which significantly reduces circuit power consumption while maintaining high accuracy.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Guibiao Liao et al.
Summary: This article proposes a novel CCFENet model for RGB-T salient object detection. The model addresses the issue of defective modalities using a cross-collaboration enhancement strategy (CCE) and aggregates multi-level complementary multi-modal features using a cross-scale cross-modal decoder (CCD). Experimental results demonstrate that CCFENet outperforms existing models on multiple datasets.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Nian Liu et al.
Summary: This paper presents a new approach to address the problem of RGB-D salient object detection by fusing attention and context from different modalities. The proposed method achieves high-order and trilinear cross-modal interaction by leveraging complementary attention cues. The paper also proposes contrast inference and selective attention mechanisms to enhance model performance, and introduces a large-scale high-quality dataset to facilitate training and evaluation of deep models.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Yu-Huan Wu et al.
Summary: This article introduces a novel network, MobileSal, for efficient RGB-D SOD, achieving good results in terms of computational efficiency and feature representation capability.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Automation & Control Systems
Jingpeng Wang et al.
Summary: Surface defect inspection of no-service rail is crucial for the safety of railway transportation. In this paper, a neural network called CLANet is proposed for defect inspection and accurate segmentation of rail surface. By integrating depth image and RGB image, the proposed method overcomes the challenges brought by irregular defect boundaries and similar foreground and background. A new cross-modal fusion strategy and dual stream decoder are introduced to enhance the detection performance. Furthermore, a new industrial dataset NEU RSDDS-AUG is built to address the scarcity of defective data. Experimental results demonstrate the effectiveness of the proposed method, which outperforms the state-of-the-art methods on various benchmark datasets.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2022)
Article
Engineering, Electrical & Electronic
Zhengyi Liu et al.
Summary: SwinNet is a model based on Swin Transformer for RGB-D and RGB-T salient object detection, combining the strengths of CNNs and Transformers, optimizing features through attention mechanism and edge information.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Information Systems
Songsong Duan et al.
Summary: This paper proposes a novel network A2TPNet for RGB-D salient object detection, which combines complementary RGB and depth modal features through alternate steered attention and trapezoidal pyramid fusion, and enhances the expressive ability of contextual semantics. Experimental results demonstrate that the proposed method outperforms 17 state-of-the-art methods on multiple datasets.
Article
Automation & Control Systems
Jie Wang et al.
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
Engineering, Electrical & Electronic
Gang Chen et al.
Summary: This paper proposes a novel RGB-T saliency object detection network, CGMDRNet, which achieves feature fusion by reducing the modality differences between the RGB and thermal features. By using different modules and strategies, the network enhances the performance and competitive with state-of-the-art models.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2022)
Article
Computer Science, Artificial Intelligence
Chang Xu et al.
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)
Article
Computer Science, Artificial Intelligence
Fushuo Huo et al.
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
Computer Science, Artificial Intelligence
Xulin Song et al.
Summary: This paper proposes a Dual Prototype Contrastive learning with Fourier Generalization (DPCFG) framework for domain adaptive Person Re-identification. The DPCFG method bridges the domain gap using the Fourier Generalization strategy and fully exploits the hard positive instances within each cluster using the Dual Prototype Contrastive learning strategy. Experimental results demonstrate that DPCFG achieves state-of-the-art UDA person ReID performance.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Lvlong Lai et al.
Summary: In this study, a simple yet effective method is proposed to tackle the problem of background ambiguities in few-shot point cloud semantic segmentation. The method incorporates the entropy of predictions on query samples into the training objective function as an additional regularization, and employs a feature transformation operation to reduce the feature differences between support and query samples. The proposed approach achieves significant performance improvement in few-shot segmentation tasks and outperforms current state-of-the-art methods across multiple task settings.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Fushuo Huo et al.
Summary: In this paper, a real-time One-stream Semantic-guided Refinement Network (OSRNet) is proposed for RGB-T SOD. The OSRNet achieves real-time speed on a single GPU through early fusion strategy and lightweight decoder, and outperforms other methods in terms of multiple evaluation metrics.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Engineering, Electrical & Electronic
Fan Sun et al.
Summary: This paper introduces a hierarchical decoding network based on Swin Transformer for red-green-blue and thermal (RGB-T) salient object detection. Compared to conventional deep convolutional neural networks, this network can better capture global information of an image and is more effective in capturing semantic associations over longer ranges.
IEEE SIGNAL PROCESSING LETTERS
(2022)
Article
Computer Science, Artificial Intelligence
Zhengzheng Tu et al.
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, Artificial Intelligence
Wujie Zhou et al.
Summary: In this study, a novel RGB-D salient object detection model using a three-input multilevel fusion network (TMFNet) is proposed, which effectively integrates RGB and depth information for improved localization of salient objects.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Wujie Zhou et al.
Summary: The article introduces a method for all-day RGB-thermal SOD, which uses infrared thermal images to supplement RGB images and improves the performance of salient object detection through adversarial learning and perceived importance fusion network.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2022)
Article
Engineering, Electrical & Electronic
Qiang Zhang et al.
Summary: This article explores RGB-T saliency detection and introduces a novel deep feature fusion network that outperforms other state-of-the-art methods by integrating multi-modality and multi-level feature fusion modules.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
Deng-Ping Fan et al.
Summary: This article makes contributions to RGB-D SOD by collecting a new SIP dataset, conducting a large-scale benchmark comparing contemporary methods, and proposing the D(3)Net model. D(3)Net outperforms prior contenders and can efficiently extract salient object masks for real scenes.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2021)
Article
Computer Science, Artificial Intelligence
Yingjie Zhai et al.
Summary: The paper introduces a novel Bifurcated Backbone Strategy Network (BBS-Net) for RGB-D salient object detection, which significantly outperforms 18 state-of-the-art models on challenging datasets. Extensive experiments demonstrate the superiority of the approach.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Engineering, Electrical & Electronic
Qinling Guo et al.
Summary: The paper introduces a two-stage fusion network (TSFNet) for integrating RGB and thermal information for RGB-T SOD. The network includes a feature-wise fusion module and a bilateral auxiliary fusion module, with multiple supervision used to further enhance SOD performance.
IEEE SIGNAL PROCESSING LETTERS
(2021)
Article
Computer Science, Artificial Intelligence
Zuyao Chen et al.
Summary: This paper proposes a novel network named DPANet, which addresses the two main challenges in RGB-D salient object detection: integrating cross-modal data complementarity and preventing contamination effects from unreliable depth maps. By introducing depth potentiality perception and gated multi-modality attention module, the proposed approach effectively models the potential of depth maps and integrates cross-modal complementarity.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Zhengzheng Tu et al.
Summary: In this study, a multi-interactive dual-decoder is proposed to mine and model the multi-type interactions for accurate RGBT SOD. The method performs well on public datasets and can handle challenging scenarios even in the presence of invalid modality.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Review
Computer Science, Software Engineering
Tao Zhou et al.
Summary: This paper provides a comprehensive survey of RGB-D based salient object detection models and related benchmark datasets, as well as discusses the challenges and open directions for future research in this field.
COMPUTATIONAL VISUAL MEDIA
(2021)
Article
Computer Science, Artificial Intelligence
Gongyang Li et al.
Summary: In this study, a novel RGB-D Salient Object Detection method, HAINet, is proposed to mitigate distractions in depth maps and highlight salient objects in RGB images through three key stages.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Information Systems
Zhengzheng Tu et al.
IEEE TRANSACTIONS ON MULTIMEDIA
(2020)
Article
Engineering, Electrical & Electronic
Jin Tang et al.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2020)
Article
Engineering, Electrical & Electronic
Liming Huang et al.
IEEE SIGNAL PROCESSING LETTERS
(2020)
Article
Computer Science, Artificial Intelligence
Hao Chen et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2020)
Article
Computer Science, Artificial Intelligence
Qiang Zhang et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2020)
Article
Computer Science, Artificial Intelligence
Hao Chen et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2019)
Article
Computer Science, Artificial Intelligence
Liang-Chieh Chen et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2018)
Article
Computer Science, Artificial Intelligence
Liangqiong Qu et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2017)
Article
Engineering, Electrical & Electronic
Runmin Cong et al.
IEEE SIGNAL PROCESSING LETTERS
(2016)