• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.10
• /
• pp.4080-4097
• /
• 2020

In recent years, convolutional neural network (CNN) has become the primary method for face detection. But its shortcomings are obvious, such as expensive calculation, heavy model, etc. This makes CNN difficult to use on the mobile devices which have limited computing and storage capabilities. Therefore, the design of lightweight CNN for face detection is becoming more and more important with the popularity of smartphones and mobile Internet. Based on the CPU real-time face detector FaceBoxes, we propose a multi-task lightweight face detector, which has low computing cost and higher detection precision. First, to improve the detection capability, the squeeze and excitation modules are used to extract attention between channels. Then, the textual and semantic information are extracted by shallow networks and deep networks respectively to get rich features. Finally, the landmark detection module is used to improve the detection performance for small faces and provide landmark data for face alignment. Experiments on AFW, FDDB, PASCAL, and WIDER FACE datasets show that our algorithm has achieved significant improvement in the mean average precision. Especially, on the WIDER FACE hard validation set, our algorithm outperforms the mean average precision of FaceBoxes by 7.2%. For VGA-resolution images, the running speed of our algorithm can reach 23FPS on a CPU device.

• International Journal of Computer Science & Network Security
• /
• v.21 no.4
• /
• pp.19-27
• /
• 2021

Tasks scheduling have been gaining attention in both industry and research. The scheduling that ensures independent task execution is critical in real-time systems. While task scheduling has gained a lot of attention in recent years, there have been few works that have been implemented into real-time architecture. The efficiency of the classical scheduling strategy in real-time systems, in particular, is still understudied. To reduce total waiting time, we apply three scheduling approaches in this paper: First In/First Out (FIFO), Shortest Execution Time (SET), and Shortest-Longest Execution Time (SLET). Experimental results have demonstrated the efficacy of the SLET in comparison with the others in most cases in a wide range of configurations.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.4
• /
• pp.1275-1292
• /
• 2021

As a fine-grained classification problem, aspect-level sentiment classification predicts the sentiment polarity for different aspects in context. To address this issue, researchers have widely used attention mechanisms to abstract the relationship between context and aspects. Still, it is difficult to effectively obtain a more profound semantic representation, and the strong correlation between local context features and the aspect-based sentiment is rarely considered. In this paper, a hybrid attention capsule network for aspect-level sentiment classification (ABASCap) was proposed. In this model, the multi-head self-attention was improved, and a context mask mechanism based on adjustable context window was proposed, so as to effectively obtain the internal association between aspects and context. Moreover, the dynamic routing algorithm and activation function in capsule network were optimized to meet the task requirements. Finally, sufficient experiments were conducted on three benchmark datasets in different domains. Compared with other baseline models, ABASCap achieved better classification results, and outperformed the state-of-the-art methods in this task after incorporating pre-training BERT.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.11
• /
• pp.39-46
• /
• 2022

In this paper, we propose a multi-task model that can simultaneously predict general-purpose tasks such as part-of-speech tagging, lemmatization, and dependency parsing using the UD Korean Kaist v2.3 corpus. The proposed model thus applies the self-attention technique of the BERT model and the graph-based Biaffine attention technique by fine-tuning the multilingual BERT and the two Korean-specific BERTs such as KR-BERT and KoBERT. The performances of the proposed model are compared and analyzed using the multilingual version of BERT and the two Korean-specific BERT language models.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.3
• /
• pp.465-472
• /
• 2020

Recently, various deep neural networks based methods have been proposed for audio event detection. In this study, we improved the performance of audio event detection by adopting an attention approach to a baseline CRNN. We applied context gating at the input of the baseline CRNN and added an attention layer at the output. We improved the performance of the attention based CRNN by using the audio data of strong labels in frame units as well as the data of weak labels in clip levels. In the audio event detection experiments using the audio data from the Task 4 of the DCASE 2018/2019 Challenge, we could obtain maximally a 66% relative increase in the F-score in the proposed attention based CRNN compared with the baseline CRNN.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.662-667
• /
• 2022

Semantic role labeling task(SRL) is to extract predicate and arguments such as agent, patient, place, time. In the previously SRL task studies, a pipeline method extracting linguistic features of sentence has been proposed, but in this method, errors of each extraction work in the pipeline affect semantic role labeling performance. Therefore, methods using End-to-End neural network model have recently been proposed. In this paper, we propose a neural network model using the Biaffine Average Attention model for SRL task. The proposed model consists of a structure that can focus on the entire sentence information regardless of the distance between the predicate in the sentence and the arguments, instead of LSTM model that uses the surrounding information for prediction of a specific token proposed in the previous studies. For evaluation, we used F1 scores to compare two models based BERT model that proposed in existing studies using F1 scores, and found that 76.21% performance was higher than comparison models.

• Science of Emotion and Sensibility
• /
• v.23 no.2
• /
• pp.35-50
• /
• 2020

This study aimed to investigate the changes of emotional perception biases according to the level of social phobia and the relationship with visuospatial attention allocation ability. This study used the Korean self-report assessment test (Korean version of social avoidance and distress, K-SAD) to measure the level of social phobia. Moreover, the emotional perception and useful field of view (UFOV) tasks were employed for measuring emotional perception biases and visuospatial attention allocation ability, respectively. A total of 118 participants participated in this experiment, and only 107 (males: 94, females: 13) data were analyzed due to the exclusion of response errors and other statistical problems. The average age of the participants used in the analysis was 21 years (SD: 3.64), and those participants were divided into three groups according to the K-SAD scores. Consequently, all experimental groups showed negative emotional perception biases in the emotional perception task, but the magnitudes of the biases of each group were not significantly different. Furthermore, the positive emotional perception biases were higher at higher levels of social phobia, which could be related to the tendency of interpreting positive stimuli negatively. In the UFOV task, the higher the level of social phobia, the lower the visuospatial attention allocation ability. These results suggest that the deterioration of visuospatial attention allocation ability potentially contributes to the increase of positive emotional perception biases by being difficult to perceive external stimuli. Thus, this paper discusses the potential contribution of visuospatial attention allocation ability to the increased perceptual biases of positive emotions as the level of social phobia increases.

• Korean Journal of Cognitive Science
• /
• v.19 no.3
• /
• pp.311-330
• /
• 2008

The effect of spatial attention gradually decreases as a function of the distance between the locus of attention and a target. According to this hypothesis, we tested the spatial-effect profile of visual attention when it operates on perception and memory. Experiment 1 measured accuracy of discriminating the color of a simultaneously masked target after presenting a pre-cue to either at the target location or away from the target (perception-intensive task). Experiment 2 measured accuracy of recognizing the color of several items at and around the pre-cued location (memory-intensive task). In the perception-intensive condition, the accuracy gradually dropped as the distance between the cue and target location increases. However, in the memory-intensive condition, subjects remembered only the item at the cued location. This suggests spatial attention in a memory-intensive process would operate on object-based representations. Experiment 2 showed the object-based effect observed in Experiment 1 can be also present in perception under a special circumstance. The results indicate that spatial attention can operate on object-based representations in a memory-intensive process whereas it flexibly can operate either on location-based or object-based representations in a perception-intensive process.

• Journal of Korean Academy of Nursing
• /
• v.20 no.2
• /
• pp.214-226
• /
• 1990

For a man to maintain attention, he needs to keep a certain level of arousal. An inordinate increase or decrease in the level of arousal eventually has a negative influence on attention. Precedent research has shown that the degree of attention changes when an experience of stress is related to anxiety resulting in a rise in arousal. This research was done to examine this hypothesis by looking at the 27 female students, 14 of whom had failed in the annual examination. The results of the investigation are as follows : The stress of failure in the examination was seen to raise the level of physiological arousal. Although pulse and electromyography showed no significant change, further inquiries should be made based on other types of methodology. In spite of the rise of arousal, the performance of selective task was degraded. This suggests those students failed to give moderate attention to given information for that kind of task. But the exact reason of that failure was not identified : that is it was difficult decide whether they gave too much attention to the anxiety brought about by stress. Performance of integral tasks, however, did not show any degradation. Judging from these results, stress seems to exert significant influence on attention in the selection of the appropriate information among the various options given. This offers an important hint in relation to the health care situation where nursing information is offered. Clients who receive nursing information in stressful situations may have difficulty in separating and selecting this helpful information from other options which they have acquired through their life experience. The content and terminology of nursing information may be strange and unintelligible to clients, although they are quite familiar and distinct to nurses. So, it is desirable for nurses to give, in addition and at the same time when nursing information is given, some certain related information as devices for selection, instead of merely giving nursing informations as such. So far it is not clear whether the concepts of information processing theory can be suitably applied to nursing. However, it is obvious, according to this research, that the quality of attention is disturbed in the stress situation. This is why further inquiries should be made into attention in practical nursing situation.

• Annals of Clinical Neurophysiology
• /
• v.8 no.1
• /
• pp.44-47
• /
• 2006

Background: The present study used magnetoencephalography (MEG) to investigate the neural substrates for modified version of Treisman's visual search task. Methods: Two volunteers who gave informed consent participated MEG experiment. One was 27- year old male and another was 24-year-old female. All were right handed. Experiment were performed using a 306-channel biomagnetometer (Neuromag LTD). There were three task conditions in this experiment. The first was searching an open circle among seven closed circles (open condition). The second was searching a closed circle among seven uni-directionally open circles (closed condition). And the third was searching a closed circle among seven eight-directionally open circles (random (closed) condition). In one run, participants performed one task condition so there were three runs in one session of experiment. During one session, 128 trials were performed during every three runs. One participant underwent one session of experiment. The participant pressed button when they found targets. Magnetic source localization images were generated using software programs that allowed for interactive identification of a common set of fiduciary points in the MRI and MEG coordinate frames. Results: In each participant we can found activations of anterior cingulate, primary visual and association cortices, posterior parietal cortex and brain areas in the vicinity of thalamus. Conclusions: we could find activations corresponding to anterior and posterior visual attention systems.