• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.1049-1052
• /
• 2019

최근 딥러닝을 이용해 정지 영상에 대한 연구 뿐만 아니라 동영상에 대한 연구들이 진행되고 있다. 본 논문에서는 동영상 딥러닝 기술에서 가장 주가 되고 있는 video action classification 에 대한 최신 기술들을 조사했다.

• Korean Journal of Adult Nursing
• /
• v.14 no.2
• /
• pp.165-173
• /
• 2002

Purpose: this study is to explore how useful ICNP nursing phenomena and actions classification is to describe the nursing problem and nursing action statements of nursing records. Method: The number of nursing phenomena statements found in this research were 323. Out of these 323, 222 statements can be fully classified, 62 statements can be partially classified, and 39 statements can not be classified at all by terms from the ICNP phenomena classification axis. Result: The number of nursing practice statements were 318, 252 of which can be fully classified, 63 statements can be partially classified, 3 statements cannot be classified at all by terms from the ICNP nursing action classification axis. Conclusions: In order to describe all the statements found in nursing records, not only new terms but also new axis need to be added to the ICNP.

• ETRI Journal
• /
• v.44 no.2
• /
• pp.327-338
• /
• 2022

Classification models for human action recognition require robust features and large training sets for good generalization. However, data augmentation methods are employed for imbalanced training sets to achieve higher accuracy. These samples generated using data augmentation only reflect existing samples within the training set, their feature representations are less diverse and hence, contribute to less precise classification. This paper presents new data augmentation and action representation approaches to grow training sets. The proposed approach is based on two fundamental concepts: virtual video generation for augmentation and representation of the action videos through robust features. Virtual videos are generated from the motion history templates of action videos, which are convolved using a convolutional neural network, to generate deep features. Furthermore, by observing an objective function of the genetic algorithm, the spatiotemporal features of different samples are combined, to generate the representations of the virtual videos and then classified through an extreme learning machine classifier on MuHAVi-Uncut, iXMAS, and IAVID-1 datasets.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.1
• /
• pp.49-55
• /
• 2019

The character dialogue created by AI is unnatural when compared with human-made dialogue, and it can not reveal the character's personality properly in spite of remarkable development of AI. The purpose of this paper is to classify characters through the linguistic style and to investigate the relation of the specific linguistic style with the personality. We analyzed the dialogues of 92 characters selected from total 60 movies categorized four movie genres, such as romantic comedy, action, comedy and horror/thriller, using Linguistic Inquiry and Word Count (LIWC), a text analysis software. As a result, we confirmed that there is a unique language style according to genre. Especially, we could find that the emotional tone than analytical thinking are two important features to classify. They were analyzed as very important features for classification as the precision and recall is over 78% for romantic comedy and action. However, the precision and recall were 66% and 50% for comedy and horror/thriller. Their impact on classification was less than romantic comedy and action genre. The characters of romantic comedy deal with the affection between men and women using a very high value of emotional tone than analytical thinking. The characters of action genre who need rational judgment to perform mission have much greater analytical thinking than emotional tone. Additionally, in the case of comedy and horror/thriller, we analyzed that they have many kinds of characters and that characters often change their personalities in the story.

• East Asian mathematical journal
• /
• v.36 no.5
• /
• pp.593-604
• /
• 2020

Let the circle act on a compact oriented manifold with a discrete fixed point set. At each fixed point, there are positive integers called weights, which describe the local action of S¹ near the fixed point. In this paper, we provide the author's original proof that only uses the Atiyah-Singer index formula for the classification of the weights at the fixed points if the dimension of the manifold is 4 and there are at most 4 fixed points, which made the author possible to give a classification for any finite number of fixed points.

• Korean Journal of Cognitive Science
• /
• v.18 no.1
• /
• pp.1-14
• /
• 2007

In a goal-oriented dialogue, speakers' intentions can be represented by domain actions that consist of pairs of a speech act and a concept sequence. Therefore, if we plan to implement an intelligent dialogue system, it is very important to correctly infer the domain actions from surface utterances. In this paper, we propose a statistical model to determine speech acts and concept sequences using conditional random fields at the same time. To avoid biased learning problems, the proposed model uses low-level linguistic features such as lexicals and parts-of-speech. Then, it filters out uninformative features using the chi-square statistic. In the experiments in a schedule arrangement domain, the proposed system showed good performances (the precision of 93.0% on speech act classification and the precision of 90.2% on concept sequence classification).

• The Journal of Korea Robotics Society
• /
• v.15 no.4
• /
• pp.404-412
• /
• 2020

Typically everyday human life tasks involve at least two people moving objects such as tables and beds, and the balancing of such object changes based on one person's action. However, many studies in previous work performed their tasks solely on robots without factoring human cooperation. Therefore, in this paper, we propose cooperative robot for table balancing using Q-learning that enables cooperative work between human and robot. The human's action is recognized in order to balance the table by the proposed robot whose camera takes the image of the table's state, and it performs the table-balancing action according to the recognized human action without high performance equipment. The classification of human action uses a deep learning technology, specifically AlexNet, and has an accuracy of 96.9% over 10-fold cross-validation. The experiment of Q-learning was carried out over 2,000 episodes with 200 trials. The overall results of the proposed Q-learning show that the Q function stably converged at this number of episodes. This stable convergence determined Q-learning policies for the robot actions. Video of the robotic cooperation with human over the table balancing task using the proposed Q-Learning can be found at http://ibot.knu.ac.kr/videocooperation.html.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.1
• /
• pp.1-9
• /
• 2013

This paper propose a human action recognition method that uses bag-of-features (BoF) based on CS-LBP (center-symmetric local binary pattern) and a spatial pyramid in addition to the random forest classifier. To construct the BoF, an image divided into dense regular grids and extract from each patch. A code word which is a visual vocabulary, is formed by k-means clustering of a random subset of patches. For enhanced action discrimination, local BoF histogram from three subdivided levels of a spatial pyramid is estimated, and a weighted BoF histogram is generated by concatenating the local histograms. For action classification, a random forest, which is an ensemble of decision trees, is built to model the distribution of each action class. The random forest combined with the weighted BoF histogram is successfully applied to Standford Action 40 including various human action images, and its classification performance is better than that of other methods. Furthermore, the proposed method allows action recognition to be performed in near real-time.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.426-432
• /
• 2022

HVAC systems play a critical role in reducing energy consumption in buildings. Integrating occupants' thermal comfort evaluation into HVAC control strategies is believed to reduce building energy consumption while minimizing their thermal discomfort. Advanced technologies, such as visual sensors and deep learning, enable the recognition of occupants' discomfort-related actions, thus making it possible to estimate their thermal discomfort. Unfortunately, it remains unclear how accurate a deep learning-based classifier is to recognize occupants' discomfort-related actions in a working environment. Therefore, this research evaluates the classification performance of occupants' discomfort-related actions while sitting at a computer desk. To achieve this objective, this study collected RGB video data on nine college students' cold discomfort-related actions and then trained a deep learning-based classifier using the collected data. The classification results are threefold. First, the trained classifier has an average accuracy of 93.9% for classifying six cold discomfort-related actions. Second, each discomfort-related action is recognized with more than 85% accuracy. Third, classification errors are mostly observed among similar discomfort-related actions. These results indicate that using human action data will enable facility managers to estimate occupants' thermal discomfort and, in turn, adjust the operational settings of HVAC systems to improve the energy efficiency of buildings in conjunction with their thermal comfort levels.

• Korean Journal of Clinical Pharmacy
• /
• v.22 no.3
• /
• pp.260-267
• /
• 2012

The objective of this study was to investigate drug classification system in Korea and other developed countries. Laws and regulations of Korea regarding the system were retrieved from sources posted in Ministry of Government Legislation. We also reviewed previous research reports performed as part of government's effort to reform the system The system in the foreign countries was retrieved from the official homepage operated by each country's government. There have been two research funded by Korean government, which strongly suggested that the system should be reformed. However, we found that the system was never reformed and still effective. Drug classification system in US and most western countries consists of two categories, i.e., prescription drugs and non-prescription drugs except UK, which classifies into three categories: Prescription Only Medicines, Pharmacy Medicines, and General Sales List Medicines. Interestingly, in Japan, non-prescription drugs are further classified into three groups: Group 1, 2, and 3. Recently, Ministry of Health and Welfare (MOHW) in Korea proposed a plan to reclassify all the approved drugs according to purportedly rational and scientific criteria. However, the plan does not include reform of the existing laws and regulations, which appears that it is just one-time action rather than a sustainable administration backed up by law. Therefore, it is recommended that Korean MOHW take appropriate action on laws and regulations with regard to the system to meet global harmonization standard.