• Journal of Korea Multimedia Society
• /
• v.2 no.3
• /
• pp.243-251
• /
• 1999

In this paper, we suggest a novel method which defines index for effectively searching MPEG video data. The traditional methods are slow because they rely on pattern matching of images between extracted shots and user's querying images. Whereas, our method uses a different approach: It transforms images of extracted PFrames into codes and uses the codes for creating index keys. To be specific, we define five features of non-context description for transparency about user's querying.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2000.04a
• /
• pp.366-369
• /
• 2000

This paper examines three algorithms to recognize speaker's emotion using the speech signals. Target emotions are happiness, sadness, anger, fear, boredom and neutral state. MLB(Maximum-Likeligood Bayes), NN(Nearest Neighbor) and HMM (Hidden Markov Model) algorithms are used as the pattern matching techniques. In all cases, pitch and energy are used as the features. The feature vectors for MLB and NN are composed of pitch mean, pitch standard deviation, energy mean, energy standard deviation, etc. For HMM, vectors of delta pitch with delta-delta pitch and delta energy with delta-delta energy are used. We recorded a corpus of emotional speech data and performed the subjective evaluation for the data. The subjective recognition result was 56% and was compared with the classifiers' recognition rates. MLB, NN, and HMM classifiers achieved recognition rates of 68.9%, 69.3% and 89.1% respectively, for the speaker dependent, and context-independent classification.

• The Korean Journal of Applied Statistics
• /
• v.22 no.6
• /
• pp.1345-1353
• /
• 2009

We introduce a simple methodology, so-called generalized gamma-polynomial approximation, based on moment-matching technique to approximate survival and hazard functions in the context of parametric survival analysis. We use the generalized gamma-polynomial approximation to approximate the density and distribution functions of convolutions and finite mixtures of random variables, from which the approximated survival and hazard functions are obtained. This technique provides very accurate approximation to the target functions, in addition to their being computationally efficient and easy to implement. In addition, the generalized gamma-polynomial approximations are very stable in middle range of the target distributions, whereas saddlepoint approximations are often unstable in a neighborhood of the mean.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.25 no.8
• /
• pp.983-991
• /
• 1988

In this paper, a new two dimensional processing method is presented, which determines the identity, position and orientation of a part. Matching between the object and model is performed in the frequency domain. The DFT of the object contour is decomposed to estimated the orientation of the object and to evaluate the similarity between the object and model. In this context, this new approach is very rbust with respect to noise and no preprocessing of the contour is required. Also, this method has many advantages over the conventional correlation technique. With only a few uniformly sampled points, this method can estimate the accurate orientation in an efficient manner even in a noisy environment.

• Journal of Information Processing Systems
• /
• v.18 no.1
• /
• pp.26-47
• /
• 2022

Chatbots were an important research subject in the past. A chatbot is a computer program or an artificial intelligence program that participates in a conversation via auditory or textual methods. As the research on chatbots progressed, some important issues regarding them changed over time. Therefore, it is necessary to review the technology with a focus on recent advancements and core research technologies. In this paper, we introduce five different chatbot technologies: natural language processing, pattern matching, semantic web, data mining, and context-aware computer. We also introduce the latest technology for the chatbot researchers to recognize the present situation and channelize it in the right direction.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2009.11a
• /
• pp.169-170
• /
• 2009

새로운 상호작용 장치와 응용 시스템에 대한 유비쿼터스 접근을 위한 요구사항이 창안되어, 사용자 상호작용은 데스크탑 너머로 이동하여 개발되는 경향이 있다. 사용되어지는 응용에서 상황은 시스템을 수행하는 활동의 총체적인 부분이 되었다. 상황 인지를 포함하는 것은 유비쿼터스 접근을 위해서 사용자에게 편리함과 효율성을 제공해 준다. 우리는 상황의 개념이 유비쿼터스 컴퓨팅을 위한 컴퓨팅 상황, 사용자 상황과 물리적 상황의 서로 다른 카테고리로 확장되어져야 한다고 믿고 있기 때문에, 전통적인 HCI 이론은 상황 인지 응용을 개발하는데 부적절하다. 이것들은 확장된 상황 정보를 잘 사용하기 위해 시스템 요구사항 분석과 설계에 있어서 새로운 패러다임을 요구한다. 본 연구에서는 성능 매칭(capability matching)에 의해서 상황 인식 요구사항 분석을 위한 방법과 목표 상황에 대한 상황 인식 속성을 매치시키는 방법을 소개한다. 이런 모델을 근거로 본 연구에서는 몇몇 공통적인 유비쿼터스 접근 상황의 설계 문제를 분석하고 유비쿼터스 관광 응용 요구사항을 고려함에 의해 상황 인지 응용을 체계적으로 적용시키는 방법을 보여준다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.646-647
• /
• 2022

Multiview stereo (MVS) 3D reconstruction of a scene from images is a fundamental computer vision problem that has been thoroughly researched in recent times. Traditionally, MVS approaches create dense correspondences by constructing regularizations and hand-crafted similarity metrics. Although these techniques have achieved excellent results in the best Lambertian conditions, traditional MVS algorithms still contain a lot of artifacts. Therefore, in this study, we suggest using a transformer network to accelerate the MVS reconstruction. The network is based on a transformer model and can extract dense features with 3D consistency and global context, which are necessary to provide accurate matching for MVS.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.6
• /
• pp.177-193
• /
• 2022

In the near future, it is expected that the use of autonomous mobility-on-demand services will increase. Considering its complicated service components, including vehicle convenience, driving and matching speed, and platform convenience, the priorities of them will need to be determined for a successful establishment. In this context, this study examined the importance rank of each service component through an online survey of potential users of autonomous mobility-on-demand services. As a result of the AHP (Analytic Hierarchy Process) with respect to the upper-level components, driving and matching speed component is selected as most important, followed by platform convenience and vehicle convenience. Mean rank analysis with respect to lower-level components showed that the in-vehicle congestion level of vehicle convenience, waiting time of driving and matching speed, and pre-booking availability of platform convenience each ranked first. Additional analysis regarding each group was conducted to establish a group-specific strategy. As a result, it would be better to focus on a vehicle than a mobile platform when designing services for the region with a high proportion of the older. Moreover, it is recommended to speed up the driving and matching speeds more than the current public transport, alleviate in-vehicle congestion, and enable the users to book the schedule in advance.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.3
• /
• pp.123-129
• /
• 2009

Ubiquitous home is emerging as the future digital home environments that provide various ubiquitous home services like u-Life, u-Health, etc. It is composed of some home appliances and sensors which are connected through wired/wireless network. Ubiquitous home services become aware of user's context with the information gathered from sensors and make home appliances adapt to the current home situation for maximizing user convenience. In these context-aware home environments, it is the one of significant research topics to predict user behaviors in order to proactively control the home environment. In this paper, we propose Multi-Level prediction algorithm for context-aware services in ubiquitous home environment. The algorithm has two phases, prediction and execution. In the first prediction phase, the next location of user is predicted using tree algorithm with information on users, time, location, devices. In the second execution phase, our table matching method decides home appliances to run according to the prediction, device's location, and user requirement. Since usually home appliances operate together rather than separately, our approach introduces the concept of mode service, so that it is possible to control multiple devices as well as a single one. We also devised some scenarios for the conceptual verification and validated our algorithm through simulations.

• Journal of Satellite, Information and Communications
• /
• v.9 no.1
• /
• pp.57-64
• /
• 2014

Recently, many human-friendly communication methods have been studied for human-machine interface(HMI) without using any physical devices. One of them is the vision-based gesture recognition that this paper deals with. In this paper, we define some gestures for interaction with objects in a predefined virtual world, and propose an efficient method to recognize them. For preprocessing, we detect and track the both hands, and extract their silhouettes from the low-resolution hand images captured by a webcam. We modeled skin color by two Gaussian distributions in RGB color space and use blob-matching method to detect and track the hands. Applying the foodfill algorithm we extracted hand silhouettes and recognize the hand shapes of Thumb-Up, Palm and Cross by detecting and analyzing their modes. Then, with analyzing the context of hand movement, we recognized five predefined one-hand or both-hand gestures. Assuming that one main user shows up for accurate hand detection, the proposed gesture recognition method has been proved its efficiency and accuracy in many real-time demos.