Article
Engineering, Electrical & Electronic
Yong Zhang, Qingqing Liu, Yujie Wang, Guangwei Yu
Summary: Channel state information (CSI)-based human activity recognition (HAR) has significant application prospects in smart homes, medical monitoring, and public security. However, the challenge lies in the fact that the collected CSI data contains both activity and environment-related information, resulting in different characteristics for the same activity at different locations. To address this issue, an Attention-based feature Fusion ACTivity recognition system (AF-ACT) is proposed, which extracts semantic activity features and temporal features from different dimensions to better characterize activities at different locations. The proposed system achieves high recognition accuracy through the fusion of semantic activity features and temporal features using an attention-based feature fusion (A-Fusion) module.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Multidisciplinary Sciences
Degang Xu, Hao Li, Ruirui Wu, Yizhi Wang, Yonghao Huang, Yaoyi Cai
Summary: This study proposes an efficient attention mechanism algorithm and a nonlinear feature fusion module for metal roll detection. A multi-scale object detection network is developed to accurately detect metal rolls. Experimental results demonstrate the effectiveness of the proposed network.
Article
Computer Science, Artificial Intelligence
Qiangxi Zhu, Zhixin Li, Wenlan Kuang, Huifang Ma
Summary: This paper proposes an efficient global channel position-aware interaction method for fine-grained visual classification. The method enhances the receptive field of global features by hierarchically grouping original features and utilizing the translation-invariant linearity and local weight sharing of convolutional networks. Same-direction location attention interaction is performed based on the global feature to capture common areas of interest. Multiple attention feature map is obtained based on the relative position interactions of the global features, and discriminative feature regions are learned and optimized for guiding the classification process. The proposed model performs well on CUB-200-2011, Stanford Cars, and FGVC Aircraft datasets.
APPLIED INTELLIGENCE
(2023)
Article
Engineering, Civil
Guiming Sun, Heng Qi, Yanming Shen, Baocai Yin
Summary: In this paper, a temporal-context-based self-attention network named TCSA-Net is proposed, which can simultaneously exploit long-and short-term movement preferences from sparse and long trajectories. The network outperforms state-of-the-art methods in terms of standard evaluation metrics, thanks to its novel two-stage self-attention architecture and multi-modal embedding layer.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Neurosciences
Angela Renton, David R. Painter, Jason B. Mattingley
Summary: The study used a neural feedback protocol to bias sensory processing, successfully influencing integrative decision-making processes.
JOURNAL OF NEUROSCIENCE
(2021)
Article
Neurosciences
Jianrong Jia, Ying Fan, Huan Luo
Summary: Recent studies show that attention operates rhythmically, sampling each location or feature alternatively over time. This study used behavioral measurements and EEG recordings to investigate the role of alpha-band neural oscillations in feature processing. The results suggest that different saliency levels of features are processed at different phases of alpha neural oscillations.
Article
Computer Science, Information Systems
Jian Shi, Ge Sun, Jinyu Zhang, Zhihui Wang, Haojie Li
Summary: In this paper, a weakly supervised attribute location module (ALM) is proposed to effectively detect facial regions with only image-level attribute labels, and improve face attribute recognition using region-based local features. Moreover, a bottom-up skip connection structure is introduced to enhance attribute-specific region location with low-level spatial information supplements. Extensive experiments demonstrate the superior performance of the proposed method on LFWA and CelebA datasets.
MULTIMEDIA SYSTEMS
(2023)
Article
Computer Science, Information Systems
Shuai Xu, Dechang Pi, Jiuxin Cao, Xiaoming Fu
Summary: The study proposes a two-stage framework consisting of a temporal base model and a location prediction model to predict user consumption locations in the future. The first stage captures user latent preference using user sentimental textual reviews and hierarchical attention mechanism, while the second stage derives consumption probability towards different locations by incorporating multifaceted context information.
INFORMATION PROCESSING & MANAGEMENT
(2021)
Article
Computer Science, Artificial Intelligence
Tao Huang, Rui Fu
Summary: A novel DFoA prediction method based on feature visualization of a deep autonomous driving model is proposed in this study, which does not require ground-truth DFoA data for training. By employing a multimodal spatiotemporal convolutional network and attention mechanism, the method accurately predicts DFoA by fusing features and utilizing ConvLSTM network for successive frames.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Zhen Wang, Shanwen Zhang, Wenzhun Huang, Jianxin Guo, Leya Zeng
Summary: The AGFE-Net proposed in this study utilizes multi-scale convolution and attention mechanisms with a global receptive field to enhance feature extraction, suppress background noise interference, and effectively fuse features of different scales in sonar image processing. Experimental results demonstrate the superiority of AGFE-Net in sonar target detection.
IEEE SENSORS JOURNAL
(2022)
Article
Engineering, Electrical & Electronic
Zhimin Tang, Guobao Xiao, Junwen Guo, Shiping Wang, Jiayi Ma
Summary: In this article, a dual-attention-based feature aggregation network is proposed for infrared and visible image fusion. The network effectively aggregates useful features and adaptively integrates meaningful features through multiple branch channel attention and global-local spatial attention. The fusion process is evaluated using multiscale structural similarity as a loss function. Extensive experiments demonstrate the superiority of the proposed network compared to state-of-the-art methods on multiple datasets.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Computer Science, Artificial Intelligence
Kangkang Ji, Pan Huang, Songhao Zhu
Summary: This paper proposes a new attention-calibration double-branch cross-domain pedestrian re-identification network, which achieves excellent performance in various environments by learning features from different domains and fusing them.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Psychology, Mathematical
Angus F. Chapman, Viola S. Stormer
Summary: Feature-based theories suggest that features not primarily attended are enhanced globally, surpassing object boundaries. Object-based theories propose that secondary, task-irrelevant features are enhanced within object boundaries.
PSYCHONOMIC BULLETIN & REVIEW
(2021)
Article
Computer Science, Information Systems
Hao Wang, Yugui Wang, Rui Cui, Yibo Han, Chaohua Yan, Menghan Niu
Summary: Crowdsourcing offers an effective way to build a location recognition image database, with rich information but susceptible to disturbances. To address this, a RCCF detection framework and VHB scheme were proposed, along with deep feature extraction, leading to superior performance compared to other state-of-the-art schemes.
Article
Automation & Control Systems
Jing Cheng, Rongjie Wang, Anhui Lin, Desong Jiang, Yichun Wang
Summary: Instance-level ship recognition (ISR) has important applications in civil and military fields. This study proposes a feature enhanced RetinaNet-based method for ISR (FERISR), which enhances the salience features of ship instances and utilizes attention-guided features to improve recognition accuracy in different scenarios.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Neurosciences
Ryan P. Silk, Hanagh R. Winter, Ouria Dkhissi-Benyahya, Carmella Evans -Molina, Alan W. Stitt, Vijay K. Tiwari, David A. Simpson, Eleni Beli
Summary: This study investigates whether diabetes affects the daily rhythm of gene expression in the retina. The results show that diabetic mice exhibited phase advancement in the expression of certain genes compared to non-diabetic mice. The study also identified oxygen-sensing mechanisms and HIF1alpha as potential upstream regulators. These findings provide important insights into the development of diabetic retinopathy.
Article
Neurosciences
Krishnamachari S. Prahalad, Daniel R. Coates
Summary: Visual stimuli presented around the time of a saccade can be perceived differently by the visual system, including a reduction in the harmful impact of flankers. This study investigated the effects of microsaccades on crowded stimuli placed 20 arc minutes from the center of gaze. The findings suggest two separate pre-saccadic benefits, one that regularizes the crowding zone and another that specifically benefits microsaccade targets surrounded by tangential flankers.
Article
Neurosciences
Chandrika Ravisankar, Christopher W. Tyler, Clifton M. Schor, Shrikant R. Bharadwaj
Summary: This study revealed that less than one-third of adults with normal binocular vision were able to successfully free-fuse random-dot image pairs and identify the embedded stereoscopic shapes. The successful participants showed a dissociation of vergence and accommodative responses, while the unsuccessful ones either exhibited strong vergence and accommodation or weak vergence and strong accommodation. Task performance of the unsuccessful cluster improved significantly with pharmacological paralysis of accommodation. A minority of participants also learned to dissociate one direction of their vergence and accommodation crosslinks with repeated free-fusion trials, optimizing their task performance.
