• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.12
• /
• pp.224-231
• /
• 2013

Defining the concept of prominent audio content as the most informative audio content from the users' perspective within a given unstructured audio segment, we propose a simple but robust intensity-based attack/release pattern features to detect the prominent audio content. We also propose a web-based annotation procedure to retrieve users' subjective perception and annotated 18 hours of video clips across various genres, such as cartoon, movie, news, etc. The experiments with a linear classification method whose models are trained for speech, music, and sound effect demonstrate promising - but varying across the genres of programs - results (e.g., 86.7% weighted accuracy for speech-oriented talk shows and 49.3% weighted accuracy for {action movies}).

• Annual Conference of KIPS
• /
• 2011.04a
• /
• pp.441-443
• /
• 2011

본 논문은 오디오 특징 분석을 기반으로 한 음악검색 방법에 대한 기술과 연구에 대한 내용이다. 본 연구에서는 크게 3가지의 주요 알고리즘을 이용하여 다 성음에서의 오디오 특징을 추출하고 3가지의 각자 다른 방식의 매칭 알고리즘을 기반으로 한 퓨전 매칭 방식을 제안한다. 오디오 특징으로는 메인 멜로디, 음악 구조를 분석한 세그먼테이션 정보를 이용한다. 본 연구에서 사용된 음악 DB는 음악 포털 서비스에서 제공하는 장르를 기반으로 한 8가지 장르에서 다양한 범위에서 2000곡을 선곡하였다. 오디오 특징 추출을 위한 알고리즘 개발과 매칭 알고리즘 개발을 위하여 음악 DB 2000곡 중 장르의 비율을 고려하여 100곡을 선정하고, 24명으로부터 1200개의 허밍을 녹음하였다. 24명중 3명은 대학에서 음악을 전공하고 나머지는 음악적 교육을 받은 경험이 없는 사람들이다. 1200개의 허밍을 분석한 결과 전체 허밍 중 60%정도가 노래의 시작 부분을 허밍하거나 노래를 불렀고, 30%정도는 하이라이트 부분을 허밍 하였다. 나머지 10%정도는 자신이 가장 자신 있는 부분을 불렀다. 이러한 분석 결과를 기반으로 가장 중요한 부분은 노래가 시작되는 부분에서의 멜로디를 정확하게 찾아내는 것이 무엇보다 중요하다는 것이다. 본 연구에서 검색결과의 평가는 MRR를 이용하여 측정하였다. MIDI DB를 사용한 경우가 다 성음에서 직접 멜로디를 추출한 경우보다 약간 성능이 우수하게 나왔으나 그 차이는 미미했다. 본 연구에서는 개발된 알고리즘을 이용하여 PC상에서 사용할 수 있는 클라이언트 프로그램과 Android app를 개발하였다.

• Annual Conference of KIPS
• /
• 2011.04a
• /
• pp.475-478
• /
• 2011

지금까지 음악 정보 검색을 위한 다양한 내용 기반 음악 검색 및 비교 방법이 제안되었다. 그런데, 기존 연구들은 질의 방식 및 검색 카테고리가 변화함에 따라 상이한 방법을 제시하고 있어 음악 검색 방법을 통합하는 데에 한계가 있다. 이러한 문제를 해결하기 위해, 본고에서는 내용 기반 음악 검색의 일반화를 위한 내용 기반 음악 요소 검색(CBMER) 방법을 제안하였다. 제안 방법에서는 확률적 은닉 성분 분석(PLCA)을 사용하여 음원을 분해하고, 각 분해 요소로부터 오디오 특성을 추출하였다. 제안 방법을 사용하여 다양한 질의 방식 및 검색 카테고리로 내용 기반 음악 요소 검색이 가능함을 보이기 위해, 남성/여성의 목소리로부터 질의를 생성하여 목소리 성별에 따른 음악을 검색하는 실험을 수행하고 그 결과를 분석하였다.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.38 no.5
• /
• pp.58-64
• /
• 2001

Recently many informations are transmitted ,md stored as video data, and they are on the rapid increase because of popularization of high performance computer and internet. In this paper, to retrieve video data, shots are found through analysis of video stream and the method of detection of key frame is studied. Finally users can retrieve the video efficiently. This Paper suggests a new feature that is robust to object movement in a shot and is not sensitive to change of color in boundary detection of shots, and proposes the characterizing value that reflects the characteristic of kind of video (movie, drama, news, music video etc,). The key frames are pulled out from many frames by using the local minima and maxima of differential of the value. After original frame(not de image) are reconstructed for key frame, indexing process is performed through computing parameters. Key frames that arc similar to user's query image arc retrieved through computing parameters. It is proved that the proposed methods are better than conventional method from experiments. The retrieval accuracy rate is so high in experiments.

• The KIPS Transactions:PartD
• /
• v.14D no.7
• /
• pp.727-732
• /
• 2007

Most of previous work on indexing and searching time series focused on the similarity matching and retrieval of one-attribute time series. However, multimedia databases such as music, video need to handle the similarity search in multi-attribute time series. The limitation of the current similarity models for multi-attribute sequences is that there is no consideration for attributes' sequences. The multi-attribute sequences are composed of several attributes' sequences. Since the users may want to find the similar patterns considering attributes's sequences, it is more appropriate to consider the similarity between two multi-attribute sequences in the viewpoint of attributes' sequences. In this paper, we propose the similarity search method based on attributes's sequences in multi-attribute time series databases. The proposed method can efficiently reduce the search space and guarantees no false dismissals. In addition, we give preliminary experimental results to show the effectiveness of the proposed method.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.760-765
• /
• 2008

The purpose of this paper is to propose a powerful alternative in designing knowledge portals using Topic Maps(TM). All information is processing for one topic, about each topics define to association and composite flow of information. and all topics about add occurrence to use in the OpenAPI. Also represent to UI of graphic for intuitional representation, and using JavaScript for Cross-Browsing. so that not using XTM of a standard Topic Maps(TM) and using JSON for a simple represent to data. The results made for intuitional process UI and extensive limits to display for new information before until now musician search system. In future positive to musician search for music.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.7
• /
• pp.285-292
• /
• 2014

A named entity recognition method is used to improve the performance of information retrieval systems, question answering systems, machine translation systems and so on. The targets of the named entity recognition are usually PLOs (persons, locations and organizations). They are usually proper nouns or unregistered words, and traditional named entity recognizers use these characteristics to find out named entity candidates. The titles of books, movies and TV programs have different characteristics than PLO entities. They are sometimes multiple phrases, one sentence, or special characters. This makes it difficult to find the named entity candidates. In this paper we propose a method to quickly extract title named entities from news articles and automatically build a named entity dictionary for the titles. For the candidates identification, the word phrases enclosed with special symbols in a sentence are firstly extracted, and then verified by the SVM with using feature words and their distances. For the classification of the extracted title candidates, SVM is used with the mutual information of word contexts.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.10C
• /
• pp.965-974
• /
• 2007

In this paper, we proposed the feature selection algorithm for multi-class genre classification. In our proposed algorithm, we developed GMM separation score based on Gaussian mixture model for measuring separability between two genres. Additionally, we improved feature subset selection algorithm based on sequential forward selection for multi-class genre classification. Instead of setting criterion as entire genre separability measures, we set criterion as worst genre separability measure for each sequential selection step. In order to assess the performance proposed algorithm, we extracted various features which represent characteristics such as timbre, rhythm, pitch and so on. Then, we investigate classification performance by GMM classifier and k-NN classifier for selected features using conventional algorithm and proposed algorithm. Proposed algorithm showed improved performance in classification accuracy up to 10 percent for classification experiments of low dimension feature vector especially.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.8
• /
• pp.697-705
• /
• 2009

An unsupervised (blind) method is proposed aiming at extracting rhythmic sources from commercial polyphonic music whose number of channels is limited to one. Commercial music signals are not usually provided with more than two channels while they often contain multiple instruments including singing voice. Therefore, instead of using conventional modeling of mixing environments or statistical characteristics, we should introduce other source-specific characteristics for separating or extracting sources in the under determined environments. In this paper, we concentrate on extracting rhythmic sources from the mixture with the other harmonic sources. An extension of nonnegative matrix factorization (NMF), which is called nonnegative matrix partial co-factorization (NMPCF), is used to analyze multiple relationships between spectral and temporal properties in the given input matrices. Moreover, temporal repeatability of the rhythmic sound sources is implicated as a common rhythmic property among segments of an input mixture signal. The proposed method shows acceptable, but not superior separation quality to referred prior knowledge-based drum source separation systems, but it has better applicability due to its blind manner in separation, for example, when there is no prior information or the target rhythmic source is irregular.