Abstract:
A method for building a database based on images, which effectively builds database and shortens a road searching time is disclosed. The method comprises steps of securing an image of a crossroad; extracting information on features from the secured image of the crossroad; storing the information on features by matching the information on features with information on a map of the crossroad. Therefore, the resource and time consumed for building the database can be reduced and easy road and street searching service is provided to users.
Abstract:
본 발명은 두 개의 디지털 비디오를 비교 검색하여, 내용 중복이 발생된 구간을 보다 신속하고 정확하게 검색할 수 있도록 하는 디지털 비디오 관리 및 검색 시스템에 관한 것으로, 입력 비디오를 복호화하고 특징점을 추출하는 비디오 특징점 추출기; 비교 대상 비디오의 특징점을 저장 및 관리하는 비디오 특징점 DB(DataBase); 및 상기 입력 비디오의 특징점과 상기 비교 대상 비디오의 특징점을 코어스 비교하여 후보군을 획득하고, 상기 후보군에 대해 파인 비교를 수행하여 내용 중복이 발생된 구간을 검출 및 통보하는 비디오 특징점 비교기를 포함할 수 있다.
Abstract:
PURPOSE: A high speed adjustment system for a digital video is provided to effectively adjust video frames by including two kinds of a key frame index table. CONSTITUTION: When a digital video is inputted, a video characteristic point extraction apparatus(100) extracts characteristic points of video frames in the digital video. A characteristic point index mapping unit(400) maps a plurality of indexes of the video characteristic points. A video characteristic point comparison apparatus(300) output video information corresponding to the adjusted index by comparing the index table of a video characteristic point database(200) with the index corresponding to the video characteristic point.
Abstract:
PURPOSE: A method and apparatus for classifying photographs using additional information is provided to regard as same person when the comparison result of face feature of a designated photograph, face feature of recommended feature, and face feature of at least one or more saved photographs are similar. CONSTITUTION: A method for classifying photographs using additional information: a step of extracting photo classification unit extracting face feature additional information from one or more saved photos; a step(220,230); a step of extracting face feature and additional information from a designated photograph; a step of registering recommended photos according to the comparison result; a step of classifying photos according to the comparison result of face feature of a designated photograph, face feature of recommended feature, and face feature of at least one or more saved photographs.
Abstract:
PURPOSE: A contents searching apparatus and method are provided to extract audio features robust to various distortion with a simple computation such as filtering. CONSTITUTION: A content searching unit extracts an audio signal from video contents or audio contents(S410). The content searching unit extracts features from an audio signal extracted through filtering(S420). The content searching unit refers to contents including the extracted feature from a database(S430). Using the audio feature robust to distortion, the content searching unit improves the accuracy of content searching.
Abstract:
본발명은정합쌍을이용한신발영상검색장치및 방법에관한것으로서, 본발명에따른정합쌍을이용한신발영상검색장치는정합쌍에대한영상변형정보를고려하여대응관계를정규화하고, 정규화된대응관계를적용하여유사이미지를검색하도록함으로써, 입력된영상에대응되는영상정보를데이터베이스로부터정확하게검색하여제공할수 있는것을특징으로한다. 본발명에따르면, 입력된영상과데이터베이스에저장된영상사이의정합쌍으로부터최적의영상변형정보를검출할수 있고, 검출된영상변형정보를토대로입력된신발영상내 다수의객체를동시검색할수 있고, 따라서전술한바와같은검색으로효율적인신발검색서비스를제공할수 있다.