Electronics and Telecommunications Research Institute

Weon Geun Oh

Keun Dong Lee

Young Ho Suh

Seung Jae Lee

Da Un Jung

Yong-Seok Seo

Dong Seok Jeong

Hyung Kwan Son

Seungjae Lee

Sang Il Na

Sang Kwang Lee

Young-Suk Yoon

Wook Ho Son

Sanghyun Joo

Ik Hwan Cho

Jong Gook Ko

Wonyoung Yoo

Sung Kwan Je

Jin Soo Seo

Won Keun Yang

Chang Joon Park

Kyung Ae Moon

Hyuk Jeong

Won Young Yoo

Su Young Chi

Wookho Son

Seon Tae Kim

Hyeon Jin Kim

Gil Haeng Lee

Young Sup Hwang

Da-Un Jung

Chi Jung Hwang

Hae Gwang Kim

A Young Cho

Hyeong Yong Jeon

Ju Kyong Jin

Ayoung Cho

Ju Kyoung Jin

Sangil Na

Sungkwan Je

Dae Geun Jang

Jun Woo Lee

An apparatus and method for searching for a building on the basis of an image and a method of constructing a building search database (DB) for image-based building search. The method includes constructing a building search DB, receiving a query image from a user terminal, detecting a region to which a building belongs in the query image, extracting features of the region detected in the query image, and searching the building search DB for a building matching the extracted features. Therefore, building search performance can be improved.

Apparatus and method for searching for building based on image and method of constructing building search database for image-based building search

A machine learning method for learning how to form bounding boxes, performed by a machine learning apparatus, includes extracting learning images including a target object among a plurality of learning images included in a learning database, generating additional learning images in which the target object is rotated from the learning images including the target object, and updating the learning database using the additional learning images.

Machine learning method and apparatus

A server for providing a city street search service includes a street information database configured to store city street images, a feature selection unit configured to select at least one feature according to a predetermined criterion when a city street image for searching and two or more features for the image are received from a user terminal, a candidate extraction unit configured to extract a candidate list of a city street image, a feature matching unit configured to match the city street image for registration included in the extracted candidate list and the at least one selected feature, and a search result provision unit configured to provide the user terminal with a result of the matching as result information regarding the city street image for searching.

Server and method for providing city street search service

An apparatus for food search service includes a food region extractor configured to perform detection in regions in an image where food is present and extract a plurality of candidate regions; a candidate region refiner configured to cluster the candidate regions into groups according to a ratio of overlap between the candidate regions; and a search result generator configured to determine a position of a food region and a food item from the grouped candidate regions.

Apparatus and method for food search service

A method for extracting features from an image for use in a computing device, the method comprising: producing Gaussian Scale Space (GSS) images in the type of a pyramid from the image inputted to the computing device; performing a Scale Normalized Laplacian Filtering on the GSS images; detecting interest points from the images that are subject to the Scale Normalized Laplacian Filtering; and extracting features of the image using the detected interest points.

Method and apparatus for extracting image feature

Provided are an apparatus and method for automatically generating a visual annotation with respect to a massive image based on a visual language. The apparatus for automatically generating a visual annotation based on a visual language includes an image input unit configured to receive an input image, an image analyzing unit configured to extract feature information of the input image received by the image input unit, a searching unit configured to search a similar image with respect to the input Image and text information included in the similar image by using the feature information extracted by the image analyzing unit, and a visual annotation configuring unit configured to configure a visual annotation with respect to the input image by using the text information searched by the searching unit.

Apparatus and method for automatically generating visual annotation based on visual language

Provided are a shoe image retrieval apparatus and method using a matching pair, which can accurately retrieve image information corresponding to an inputted image from the database and provide the retrieved image information, by normalizing a correspondence relation in consideration of geometric image transformation about the matching pair and allowing a similar image to be retrieved by applying the normalized correspondence relation. It is possible to detect optimum geometric image transformation from a matching pair between the inputted shoe image and the image stored in the database and simultaneously retrieve a plurality of objects in the inputted shoe image based on the detected geometric image transformation, thereby providing an efficient shoe retrieval service.

Shoe image retrieval apparatus and method using matching pair

A system for managing and searching for a digital video includes: a video feature point extraction unit decoding an input video and extracting a feature point; a video feature point database (DB) storing and managing a feature point of a video to be compared (i.e., a comparison target video), and a video feature point comparison unit coarsely comparing the feature point of the input video and that of the comparison target video to acquire a candidate group, and minutely comparing the candidate group to detect a content repeated section, and informing a user of the content repeated section.

Digital video managing and searching system

An image searching method includes: resizing an input image in question; generating 3D image identifiers for the resized input image in question; and performing an image search for the input image in question by using the 3D image identifiers. Said resizing an input image includes: extracting a black-and-white image and a color image from the input image in question; resizing the black-and-white image; and resizing the color image. Said generating 3D image identifiers includes: extracting an MGST feature of the input image in question; extracting an angular partition feature of the input image in question; and extracting a color feature of the input image in question. Further, said performing an image search includes: calculating similarity between representative colors of the input image in question and a reference image; and if the similarity is above a predetermined level, matching the 3D image identifiers for the two images.

Image searching method and apparatus

A method for managing digital videos includes extracting a feature of an input digital video, searching for candidate features having similarity to the extracted feature, and comparing the extracted feature with each of the candidate features to select at least one of the candidate features similar to the extracted feature and provide information on a digital video corresponding to the selected candidate feature.

System and method for comparing an input digital video to digital videos using extracted and candidate video features

An image signature creating method includes: creating a map by partitioning a still image using rings defined by concentric circles radially spaced apart from each other by a predetermined interval and radial lines circumferentially spaced apart from each other by a predetermined angle, the center of the still image being that of the concentric circles; and creating an image signature from the created map on the basis of distribution of pixels in regions defined by the rings and the radial lines. Accordingly, distribution of illegal contents is prevented and interrupted by search for and discrimination of a still image illegally deformed by a copying of or an intended attack on the still image.

Image signature creating method and apparatus for discriminating still images

An index database creating apparatus includes a feature extracting unit for extracting features from a content; an index creating unit for creating feature indices of the extracted features; an index selector for selecting one or more of the feature indices based on frequency-based importance levels thereof; and a feature index database for storing therein the selected feature indices along with locations thereof in the content. The index selector obtains the frequency-based importance levels based on frequency distributions of the feature indices in the content and in other contents to be stored in the feature index database.

Index database creating apparatus and index database retrieving apparatus

A system and method for identifying an image based on singular value decomposition is provided. The system includes: a feature point extracting module for extracting at least one of feature points from an input image using strength of a Gaussian curvature of each pixel from the input image; an identifier detecting module for detecting a local identifier based on SVD (singular value decomposition) for blocks adjacent to the extracted feature points; and a matching module for determining whether an image is duplicated or not by comparing the detected local identifier from the identifier detecting module with an image database storing at least one of images.

System and method for identifying image based on singular value decomposition and feature point

Provided is a method of selecting test stimuli for use in evaluating performance of a video watermarking algorithm, including the steps of: calculating an image complexity rate (ICR) for each of a plurality of test stimulus candidates and classifying them into L (where L is an integer more than or equal to 2) levels based on the calculated ICR values; calculating an energy distribution rate (EDR) for each of the plurality of test stimulus candidates, and classifying them into M (where M is an integer more than or equal to 2) levels based on the calculated EDR values; calculating a motion rate (MR) for each of the plurality of test stimulus candidates and classifying them into N (where N is an integer more than or equal to 2) levels of the calculated MR values; and selecting at least one test stimulus from each of the L levels, M levels, the N levels of the test stimuli, respectively.

Method and systems for selecting test stimuli for use in evaluating performance of video watermarking methods

In an apparatus for embedding digital watermarks into an image, a watermark embedding region extraction unit extracts an image region to contain watermarks from a wavelet transformed entire image. A watermark generation unit generates a watermark data row to be embedded according to a user's private key. A normalization operation unit normalizes the quantized data value by using a quantization unit into a normalized data value. An inverse quantization inversely quantizes and outputs the normalized data value. A watermark embedment unit embeds a watermark-embedded data value as a corresponding pixel data value into the watermark embedding region, wherein the watermark-embedded data value is obtained by subtracting, from the inversely quantized data value, the watermark data value added prior to the quantization.

Growing community of inventors

Daejeon, South Korea

Weon Geun Oh