• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권12호
• /
• pp.3976-3990
• /
• 2022

Currently, cross-attention is widely used in Siamese trackers to replace traditional correlation operations for feature fusion between template and search region. The former can establish a similar relationship between the target and the search region better than the latter for robust visual object tracking. But existing trackers using cross-attention only focus on rich semantic information of high-level features, while ignoring the appearance information contained in low-level features, which makes trackers vulnerable to interference from similar objects. In this paper, we propose a Multi-level Cross-attention Siamese network(MCSiam) to aggregate the semantic information and appearance information at the same time. Specifically, a multi-level cross-attention module is designed to fuse the multi-layer features extracted from the backbone, which integrate different levels of the template and search region features, so that the rich appearance information and semantic information can be used to carry out the tracking task simultaneously. In addition, before cross-attention, a target-aware module is introduced to enhance the target feature and alleviate interference, which makes the multi-level cross-attention module more efficient to fuse the information of the target and the search region. We test the MCSiam on four tracking benchmarks and the result show that the proposed tracker achieves comparable performance to the state-of-the-art trackers.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제18권1호
• /
• pp.1-14
• /
• 2024

Time-series forecasting is extensively used in the actual world. Recent research has shown that Transformers with a self-attention mechanism at their core exhibit better performance when dealing with such problems. However, most of the existing Transformer models used for time series prediction use the traditional encoder-decoder architecture, which is complex and leads to low model processing efficiency, thus limiting the ability to mine deep time dependencies by increasing model depth. Secondly, the secondary computational complexity of the self-attention mechanism also increases computational overhead and reduces processing efficiency. To address these issues, the paper designs an efficient multi-layer attention-based time-series forecasting model. This model has the following characteristics: (i) It abandons the traditional encoder-decoder based Transformer architecture and constructs a time series prediction model based on multi-layer attention mechanism, improving the model's ability to mine deep time dependencies. (ii) A cross attention module based on cross attention mechanism was designed to enhance information exchange between historical and predictive sequences. (iii) Applying a recently proposed sparse attention mechanism to our model reduces computational overhead and improves processing efficiency. Experiments on multiple datasets have shown that our model can significantly increase the performance of current advanced Transformer methods in time series forecasting, including LogTrans, Reformer, and Informer.

• 한국소음진동공학회논문집
• /
• 제27권2호
• /
• pp.155-161
• /
• 2017

When a military vehicle is driven on a cross-country road, the occupants are exposed to vibration at a body resonance. In case that the exposure continues for too long period, the attention ability of the occupant could be affected by the vibration exposure. In the study, it was experimentally tried to find if there is a correlation between degradation of attention and vibration exposure. Two kinds of test among various psychological attention tests were employed, which were selected with considering a situation of carrying out military mission on vehicle. At the result, the searching test for controlled attention showed significant degradation in the accuracy and performance time in case of exposure at the vibration. And the attention degradation appeared to be greater when the vibration exposure increases. The dual task test for divided attention showed the difference between vibration and non-vibration condition, but showed it is insignificant for the attention to degrade by increasing exposure time.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.2115-2127
• /
• 2021

Although significant progress has been made in synthesizing visually realistic face images by Generative Adversarial Networks (GANs), there still lacks effective approaches to provide fine-grained control over the generation process for semantic facial attribute editing. In this work, we propose a novel cross channel self-attention based generative adversarial network (CCA-GAN), which weights the importance of multiple channels of features and archives pixel-level feature alignment and conversion, to reduce the impact on irrelevant attributes while editing the target attributes. Evaluation results show that CCA-GAN outperforms state-of-the-art models on the CelebA dataset, reducing Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) by 15~28% and 25~100%, respectively. Furthermore, visualization of generated samples confirms the effect of disentanglement of the proposed model.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제12권4호
• /
• pp.324-328
• /
• 2012

With the recent advances in information retrieval, cross-media retrieval has been attracting lot of attention, but several issues remain problems such as constructing effective correlations, calculating similarity between different kinds of media objects. To gain better cross-media retrieval performance, it is crucial to mine the semantic correlations among the heterogeneous multimedia data. This paper introduces a new method for cross-media retrieval which uses ontology to organize different media objects. The experiment results show that the proposed method is effective in cross-media retrieval.

• Communications for Statistical Applications and Methods
• /
• 제12권2호
• /
• pp.413-423
• /
• 2005

Consider the problem of estimating the underlying regression function from a set of noisy data which is contaminated by a long tailed error distribution. There exist several robust smoothing techniques and these are turned out to be very useful to reduce the influence of outlying observations. However, no matter what kind of robust smoother we use, we should choose the smoothing parameter and relatively less attention has been made for the robust bandwidth selection method. In this paper, we adopt the idea of robust location parameter estimation technique and propose the robust cross validation score functions.

• Physical Therapy Rehabilitation Science
• /
• 제1권1호
• /
• pp.13-16
• /
• 2012

Objective: Electroencephalogram (EGG)-feedback is a training procedure aimed at altering brain activity, and is used as a treatment for disorders like attention. The purpose of this study was to determine the effects of external focus of attention by EGG-feedback on balance in young adults. Design: Cross-sectional study. Methods: Subject were students in Sahmyook University. Fifty young adults in their twenties and thirties. Subjects were performed both with and without external focus of attention by EEG-feedback on the posture of standing and tandem standing. Participants were educated effort to maintain static posture when they were under internal focus of attention. Good Balance System was used for measurement of postural consistency upon the following force platforms. Results: Body sway decreased significantly both normal standing and tandem standing with external focus of attention by EEG-feedback (p<0.05). Conclusions: The results demonstrate that the benefits of an external attentional focus are generalizable to young adults. The external focus of attention outperformed the internal focus of attention on the postural balance (p<0.05). It is showed that external focus of attention significant effects on balance by revoked automatic postural control of movement. Furthermore balance might be improved by training with an external focus. Further study is required to develop for training as a method of preventing fall in elderly peoples.

• 정보와 통신
• /
• 제24권2호
• /
• pp.64-73
• /
• 2007

Recently, cross-layer design approach has been greatly attracting researchers' attention as an alternative for improving the performance of wireless data networks. The main reason why cross-layer approaches are particularly well suited for wireless networks is that there exists direct coupling between physical layer and upper layers. Therefore, with cross-layer approach, the protocol designers try to exploit the interaction between layers and promote adaptability at all layers, based on information exchange between layers. In this article we focus on the cross-layer engineering for high data-rate mobile Internet services through cellular networks. First, the general considerations in cross-layer engineering are outlined. Then, we discuss the common approach in literatures, which mainly deals with adaptability in physical and medium access control layer. Finally, we show that the cross-layer engineering taking account of all layers is more adequate for the mobile Internet services cellular network.

• 대한가정학회지
• /
• 제43권3호
• /
• pp.221-231
• /
• 2005

This study investigated the cross-cultural differences in temperament among Chinese, Korean-Chinese, and Korean children. Subjects were 1,046(258 Korean-Chinese, 290 Chinese, and 498 Korean) 4th graders from Yanji in China and Seoul and Ulsan in Korea. The short form of the Early Adolescent Temperament Questionnaire Scale-Revised containing 10 scales of 54 items was translated into Korean for Korean and Korean-Chinese children and into Chinese for Chinese children and was administered to children via home-room teachers. Chinese children rated themselves higher on activation control, attention, and pleasure- and perceptual-sensitivity subscales than other groups. On the contrary, Chinese-Korean children rated themselves higher on the affiliation subscale and Korean children rated themselves higher on the aggression subscale than other groups. There were larger gender differences among Chinese and Korean-Chinese than among Korean children: Gender difference was found in activation control, aggression, and attention subscales among Chinese children and in affiliation, aggression, attention and fear subscales among Korean-Chinese. There was only significant difference in pleasure sensitivity among Korean. Rapid westernization in Korea seems to be responsible for this result. Since there were significant differences of temperament between Korean and Korean-Chinese, and between Chinese and Korean-Chinese, it was interpreted that children's temperament is influenced by both their environment and genetic endowment.

• 대한물리치료과학회지
• /
• 제29권3호
• /
• pp.29-47
• /
• 2022

Purpose: This study was conducted to find out the mediated effects of smartphone overindulgence in the relationship between attention and spinal health of adolescents and to find out how cognitive strength moderates these relationships. Design: Cross-sectional study. Methods: All subjects measured attention, cognition strength, spinal angle, and smartphone overindulgence by measuring the ratio of SMR and mid-beta power to the power of the brain wave, amplitude variation of Cognitive Gamma-Peak, forward slope angle of the cervical spine, and the scale of diagnosis of smartphone addiction. Results: As a result, it shows a significant correlation between Attention, spinal angle, cognitive strength, and smartphone overindulgence(p<.05). In addition, attention was significantly mediated on the spinal angle through smartphone overindulgence(p<.001), and cognitive strength was significantly mediated in relation to smartphone overindulgence and spinal angle(p<.05). Conclusion: This result indicates that the mediation effect of overindulgence of a Smartphone is mediated according to its cognitive strength in relation to its effect of attention on the spine angle.