• The Journal of the Acoustical Society of Korea
• /
• v.23 no.2E
• /
• pp.56-67
• /
• 2004

Despite abundant research outcomes of blind source separation (BSS) in many types of simulated environments, their performances are still not satisfactory to be applied to the real environments. The major obstacle may seem the finite filter length of the assumed mixing model and the nonlinear sensor noises. This paper presents a two-step speech enhancement method with multiple microphone inputs. The first step performs a frequency-domain BSS algorithm to produce multiple outputs without any prior knowledge of the mixed source signals. The second step further removes the remaining cross-channel interference by a spectral cancellation approach using a probabilistic source absence/presence detection technique. The desired primary source is detected every frame of the signal, and the secondary source is estimated in the power spectral domain using the other BSS output as a reference interfering source. Then the estimated secondary source is subtracted to reduce the cross-channel interference. Our experimental results show good separation enhancement performances on the real recordings of speech and music signals compared to the conventional BSS methods.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.2
• /
• pp.79-88
• /
• 2021

In this paper, we propose a API Gateway technique for the expansion of human resource management module, one of the ERP functions. The institution has introduced ERP (Enterprise Resource Planning) based on its efforts to transform all human and physical resources into business competitiveness and its response to the digital knowledge informatization environment, and listed it as multiple success factors. Human resource management is one of the factors that have been dealt with. However, ERP's Human Resources Management Module remains in the role of functional personnel management. How to utilize human resources begins with navigating and recognizing human resources. The proposed API Gateway technique leverages blockchain networks to design and implement APIs for human resource sharing and navigation, including the possibility of extending ERP's human resource management module. Secondly, it was designed and implemented using a smart contract that behaves like an API for preventing information forgery. The proposed method will not only be used as a tool that can actively utilize human resources, but will also be a complete resource for utilizing big data technology.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.1
• /
• pp.67-71
• /
• 2019

With continued technology innovation in the fields of computer graphics (CG) and virtual reality (VR), digital animated avatars (or virtual humans) are evolving into ones that are more interactive at a suitable location such as museum, airport, and shopping mall. Specially, the form of the avatar (or the virtual human) realistically need to be expressed in a way that matches the users' physical space. In recent many researches, the form of virtual human has been expressed as mixed-reality human (MRH)-the virtual human combines with the physicality as the real part. In this paper, we propose to carry out a study comparing various MRH on co-presence and emotional response in two-typed virtual humans depending on how many actual parts are included: (1) (Level 1) small parts in the virtual human combined virtual components (e.g., the head only) and (2) (Level 2) large parts in the virtual human with the physicality as the real part such as head, arms, and upper body). We report on the implemented results of our virtual humans and experimental results on co-presence and emotional response.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.6
• /
• pp.2767-2780
• /
• 2016

Video-based human-activity recognition has become increasingly popular due to the prominent corresponding applications in a variety of fields such as computer vision, image processing, smart-home healthcare, and human-computer interactions. The essential goals of a video-based activity-recognition system include the provision of behavior-based information to enable functionality that proactively assists a person with his/her tasks. The target of this work is the development of a novel approach for human-activity recognition, whereby human-body-joint features that are extracted from depth videos are used. From silhouette images taken at every depth, the direction and magnitude features are first obtained from each connected body-joint pair so that they can be augmented later with motion direction, as well as with the magnitude features of each joint in the next frame. A generalized discriminant analysis (GDA) is applied to make the spatiotemporal features more robust, followed by the feeding of the time-sequence features into a Hidden Markov Model (HMM) for the training of each activity. Lastly, all of the trained-activity HMMs are used for depth-video activity recognition.

• The Journal of Korea Robotics Society
• /
• v.14 no.1
• /
• pp.22-30
• /
• 2019

For effective human-robot interaction, robots need to understand the current situation context well, but also the robots need to transfer its understanding to the human participant in efficient way. The most convenient way to deliver robot's understanding to the human participant is that the robot expresses its understanding using voice and natural language. Recently, the artificial intelligence for video understanding and natural language process has been developed very rapidly especially based on deep learning. Thus, this paper proposes robot vision to audio description method using deep learning. The applied deep learning model is a pipeline of two deep learning models for generating natural language sentence from robot vision and generating voice from the generated natural language sentence. Also, we conduct the real robot experiment to show the effectiveness of our method in human-robot interaction.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.482-485
• /
• 2021

최근 지식 그래프를 확장하기 위해 많은 연구가 진행되고 있다. 지식 그래프를 확장하기 위해서는 relation을 기준으로 entity의 방향성을 고려하는 것이 매우 중요하다. 지식 그래프를 확장하기 위한 대표적인 연구인 관계 추출은 문장과 2개의 entity가 주어졌을 때 relation을 예측한다. 최근 사전학습 언어모델을 적용하여 관계 추출에서 높은 성능을 보이고 있지만, entity에 대한 방향성을 고려하여 relation을 예측하는지 알 수 없다. 본 논문에서는 관계 추출에서 entity의 방향성을 고려하여 relation을 예측하는지 실험하기 위해 문장 수준의 Adversarial Attack과 단어 수준의 Sequence Labeling을 적용하였다. 또한 관계 추출에서 문장에 대한 이해를 높이기 위해 BERT모델을 적용하여 실험을 진행하였다. 실험 결과 관계 추출에서 entity에 대한 방향성을 고려하지 않음을 확인하였다.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.9-13
• /
• 2023

본 논문에서는 대규모 언어모델의 검색 기반 답변 생성능력을 평가하는 새로운 한국어 벤치마크, KFREB(Korean Fictional Retrieval Evaluation Benchmark)를 제안한다. KFREB는 모델이 사전학습 되지 않은 허구의 정보를 바탕으로 검색 기반 답변 생성 능력을 평가함으로써, 기존의 대규모 언어모델이 사전학습에서 보았던 사실을 반영하여 생성하는 답변이 실제 검색 기반 답변 시스템에서의 능력을 제대로 평가할 수 없다는 문제를 해결하고자 한다. 제안된 KFREB는 검색기반 대규모 언어모델의 실제 서비스 케이스를 고려하여 장문 문서, 두 개의 정답을 포함한 골드 문서, 한 개의 골드 문서와 유사 방해 문서 키워드 유무, 그리고 문서 간 상호 참조를 요구하는 상호참조 멀티홉 리즈닝 경우 등에 대한 평가 케이스를 제공하며, 이를 통해 대규모 언어모델의 적절한 선택과 실제 서비스 활용에 대한 인사이트를 제공할 수 있을 것이다.

• Asia pacific journal of information systems
• /
• v.10 no.1
• /
• pp.165-187
• /
• 2000

The computer game market has grown rapidly with numerous games produced all over the world. Most games have been developed to make gamers have fun while playing the games. However, there has been little research to address the elements of games that create the perception of being fun. The objectives of this research are to focus on which features provide fun, and then analyze these aspects both qualitatively and quantitatively. This study, through surveys with game players and developers, provides several inputs regarding what it is, that makes certain computer games fun. There are many common characteristics which fun games share, and through grouping and organizing these traits, then compiling the data for use in an AHP(Analytic Hierarchy Process), we measured the disparity in the 'fun' perception between game developer and game user.

• Journal of the Ergonomics Society of Korea
• /
• v.19 no.1
• /
• pp.77-90
• /
• 2000

The aim of this study is to develop and evaluate an efficient camera calibration method for vision-based head tracking. Tracking head movements is important in the design of an eye-controlled human/computer interface. A vision-based head tracking system was proposed to allow the user's head movements in the design of the eye-controlled human/computer interface. We proposed an efficient camera calibration method to track the 3D position and orientation of the user's head accurately. We also evaluated the performance of the proposed method. The experimental error analysis results showed that the proposed method can provide more accurate and stable pose (i.e. position and orientation) of the camera than the conventional direct linear transformation method which has been used in camera calibration. The results of this study can be applied to the tracking head movements related to the eye-controlled human/computer interface and the virtual reality technology.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1764-1768
• /
• 2004

In the near future, robots will be used for the personal use. To provide useful services to humans, it will be necessary for robots to understand human intentions. Consequently, the development of emotional interfaces for robots is an important expansion of human-robot interactions. We designed and developed an intelligent emotional interface for the robot, and applied the interfaces to our humanoid robot, AMIET. Subsequent human-robot interaction demonstrated that our intelligent emotional interface is very intuitive and friendly