• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.4026-4040
• /
• 2019

In this paper, we present an approach for detecting slide transitions in lecture videos by introducing the spatio-temporal residual networks. Given a lecture video which records the digital slides, the speaker, and the audience by multiple cameras, our goal is to find keyframes where slide content changes. Since temporal dependency among video frames is important for detecting slide changes, 3D Convolutional Networks has been regarded as an efficient approach to learn the spatio-temporal features in videos. However, 3D ConvNet will cost much training time and need lots of memory. Hence, we utilize ResNet to ease the training of network, which is easy to optimize. Consequently, we present a novel ConvNet architecture based on 3D ConvNet and ResNet for slide transition detection in lecture videos. Experimental results show that the proposed novel ConvNet architecture achieves the better accuracy than other slide progression detection approaches.

• Journal of Information Processing Systems
• /
• v.19 no.6
• /
• pp.830-841
• /
• 2023

Tokenization is the process of segmenting the input text into smaller units of text, and it is a preprocessing task that is mainly performed to improve the efficiency of the machine learning process. Various tokenization methods have been proposed for application in the field of natural language processing, but studies have primarily focused on efficiently segmenting text. Few studies have been conducted on the Korean language to explore what tokenization methods are suitable for document classification task. In this paper, an exploratory study was performed to find the most suitable tokenization method to improve the performance of a representative spatio-temporal document classifier in Korean. For the experiment, a convolutional neural network model was used, and for the final performance comparison, tasks were selected for document classification where performance largely depends on the tokenization method. As a tokenization method for comparative experiments, commonly used Jamo, Character, and Word units were adopted. As a result of the experiment, it was confirmed that the tokenization of word units showed excellent performance in the case of representative spatio-temporal document classification task where the semantic embedding ability of the token itself is important.

• Journal of KIISE
• /
• v.43 no.12
• /
• pp.1365-1375
• /
• 2016

One of the most important capabilities for autonomous service robots working in living environments is to recognize and understand the correct context in dynamically changing environment. To generate high-level context knowledge for decision-making from multiple sensory data streams, many technical problems such as multi-modal sensory data fusion, uncertainty handling, symbolic knowledge grounding, time dependency, dynamics, and time-constrained spatio-temporal reasoning should be solved. Considering these problems, this paper proposes an effective dynamic context management and spatio-temporal reasoning method for intelligent service robots. In order to guarantee efficient context management and reasoning, our algorithm was designed to generate low-level context knowledge reactively for every input sensory or perception data, while postponing high-level context knowledge generation until it was demanded by the decision-making module. When high-level context knowledge is demanded, it is derived through backward spatio-temporal reasoning. In experiments with Turtlebot using Kinect visual sensor, the dynamic context management and spatio-temporal reasoning system based on the proposed method showed high performance.

• Journal of Preventive Medicine and Public Health
• /
• v.54 no.5
• /
• pp.301-308
• /
• 2021

Objectives: Despite its advantages, it is not yet common practice in Korea for researchers to investigate disease associations using spatio-temporal analyses. In this study, we aimed to review health-related epidemiological research using spatio-temporal analyses and to observe methodological trends. Methods: Health-related studies that applied spatial or spatio-temporal methods were identified using 2 international databases (PubMed and Embase) and 4 Korean academic databases (KoreaMed, NDSL, DBpia, and RISS). Two reviewers extracted data to review the included studies. A search for relevant keywords yielded 5919 studies. Results: Of the studies that were initially found, 150 were ultimately included based on the eligibility criteria. In terms of the research topic, 5 categories with 11 subcategories were identified: chronic diseases (n=31, 20.7%), infectious diseases (n=27, 18.0%), health-related topics (including service utilization, equity, and behavior) (n=47, 31.3%), mental health (n=15, 10.0%), and cancer (n=7, 4.7%). Compared to the period between 2000 and 2010, more studies published between 2011 and 2020 were found to use 2 or more spatial analysis techniques (35.6% of included studies), and the number of studies on mapping increased 6-fold. Conclusions: Further spatio-temporal analysis-related studies with point data are needed to provide insights and evidence to support policy decision-making for the prevention and control of infectious and chronic diseases using advances in spatial techniques.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.1B
• /
• pp.13-20
• /
• 2011

The spatio-temporal pattern in precipitation is a significant element in defining characteristics of precipitation. In this study, a new scheme on regionalization utilizing temporal information was introduced on the basis of existing approaches that is mainly based on simple moments of data and geographical information. Given the identified spatio-temporal pattern, this study was extended to characterize regional pattern of annual maximum rainfall over Korea. We have used circular statistics to characterize the temporal distribution on the precipitation, and the circular statistics allow us to effectively assess changes in timing of the extreme rainfall in detail. In this study, a modified K-means method was incorporated with derived temporal characteristics of extreme rainfall in order to better characterize hydrologic pattern for regional frequency analysis. The extreme rainfall was reasonably separated into five categories that considered most attributes in both quantitative and temporal changes in extremes. The results showed that the proposed approach is a promising approach for regionalization in term of physical understanding of extreme rainfall.

• Journal of KIISE
• /
• v.43 no.9
• /
• pp.963-973
• /
• 2016

Providing a semantic knowledge system using media ontologies requires not only conventional axiom reasoning but also knowledge extension based on various types of reasoning. In particular, spatio-temporal information can be used in a variety of artificial intelligence applications and the importance of spatio-temporal reasoning and expression is continuously increasing. In this paper, we append the LOD data related to the public address system to large-scale media ontologies in order to utilize spatial inference in reasoning. We propose an RDFS/Spatial inference system by utilizing distributed memory-based framework for reasoning about large-scale ontologies annotated with spatial information. In addition, we describe a distributed spatio-temporal SPARQL parallel query processing method designed for large scale ontology data annotated with spatio-temporal information. In order to evaluate the performance of our system, we conducted experiments using LUBM and BSBM data sets for ontology reasoning and query processing benchmark.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.65 no.10
• /
• pp.1731-1737
• /
• 2016

In this paper, a method is proposed to recognize human actions as nonverbal expression; the proposed method is composed of two steps which are action representation and action recognition. First, MHI(Motion History Image) is used in the action representation step. This method includes segmentation based on depth information and generates spatio-temporal templates to describe actions. Second, CNN(Convolution Neural Network) which includes feature extraction and classification is employed in the action recognition step. It extracts convolution feature vectors and then uses a classifier to recognize actions. The recognition performance of the proposed method is demonstrated by comparing other action recognition methods in experimental results.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10c
• /
• pp.614-616
• /
• 1999

이질적인 환경(Heterogoneous Environments)에서 실시간 멀티미디어 서비스가 확산됨에 따라 비디오 데이터로 인한 트래픽이 통신망을 오가는 트래픽의 대부분을 차지하게 되었다. 이에 통신망 지원의 효율화를 위해서 사용자의 통신망 환경을 고려한, 비디오 데이터의 Layered coding과 스케러블 전송이 필요하다. 즉, 낮은 QoS 환경인 경우 일부분만을 전송하고, 높은 QoS 환경인 경우에 모두 전송하는 것이다. 이를 위해 본 논문에서는 MPEG 비디오 데이터 스케일러블 전송에 맞게 초당 디스플레이하는 프레임 수를 동적으로 조정하는 기법과 화질을 조정할 수 있는 SpatioTemporal Layered Coding 기법을 제안한다. 본 기법을 이용하면 사용자는 자신의 통신망 환경과 단말기 성능에 맞는 비디오 데이터를 수신 받아 재생할 수 있다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.10A
• /
• pp.775-779
• /
• 2003

A spatio-temporal searcher structure for 3rd generation W-CDMA systems is proposed to enhance the detection capability of the multi-path searcher for the desired signal. This searcher employs the spatio-temporal signal structure to search for newly emerging multipath signals. The proposed multi-path searcher provides better detection capability andthus reduces the mean acquisition time. The detection and false alarm probabilities of new and conventional schemes are calculated and numerical examples of mean acquisition time are given thereafter.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2004.05b
• /
• pp.454-458
• /
• 2004

A new algorithm for the rate control of videos considering the sensitivity of the human visual system is presented. The method adopts the three-step structure of MPEG2 TM5 rate control, and a new measure of macroblock activity based on spatio-temporal sensitivity is introduced for the adaptive quantization. Experimental results show that the spatio-temporal activity measure outperforms the spatial activity of TM5 in picture quality.