公开(公告)号：KR1020100073080A

公开(公告)日：2010-07-01

申请号：KR1020080131665

申请日：2008-12-22

Applicant: 한국전자통신연구원

Inventor： 최윤석 ,  박정철 ,  추창우 ,  이지형 ,  구본기 ,  김진서 ,  이승욱 ,  김호원 ,  황본우 ,  이영직

IPC: H04N13/00

CPC classification number: H04N5/23203 ,  H04N5/2627 ,  H04N5/2628

Abstract: PURPOSE: A method and an apparatus for representing a motion control camera effect based on synchronized multiple images are provided to simplify a system by performing various functions such as motion camera effect processing for a free vision moving object. CONSTITUTION: An image acquisition unit(722) acquires an image from cameras(711,712), and a geometric correction unit(724) analyzes the camera position information from an acquired image, and a color correction unit(726) corrects the colors of a multi camera image. A foreground and background separating unit(728) separates a foreground image and a background image from the acquired or corrected image. An image transmission unit(750) outputs the processed image.

Abstract translation: 目的：提供一种用于表示基于同步多个图像的运动控制相机效果的方法和装置，以通过执行诸如用于自由视觉移动物体的运动相机效果处理的各种功能来简化系统。 构成：图像获取单元（722）从摄像机（711,712）获取图像，并且几何校正单元（724）从获取的图像分析摄像机位置信息，颜色校正单元（726）校正多个 相机图像。 前景和背景分离单元（728）从获取的或校正的图像分离前景图像和背景图像。 图像传送单元（750）输出处理过的图像。

公开(公告)号：KR1020100072846A

公开(公告)日：2010-07-01

申请号：KR1020080131373

申请日：2008-12-22

Applicant: 한국전자통신연구원

Inventor： 황본우 ,  김갑기 ,  이지형 ,  김진서 ,  이승욱 ,  추창우 ,  김호원 ,  구본기 ,  이영직

IPC: H04N5/265

CPC classification number: G06T7/11 ,  G06K9/6212 ,  G06K9/6268 ,  G06T7/194 ,  G06T2207/20021

Abstract: PURPOSE: An apparatus and a method for image segmentation are provided to segment a foreground region and a background region from an inputted image including a camera sensor noise by using color distribution and binary classifier. CONSTITUTION: A binary classifier-based image segment unit(120) designates a region in an initial segment image, and generates a new segment image through an appointed area and a binary classifier. The image segment unit calculates a histogram for a background region of a new segment image. The image segment unit updates a time-spatial image block histogram of a background model generator through the sum of the weighted sum of a time-spatial image block.

Abstract translation: 目的：提供一种用于图像分割的装置和方法，用于通过使用颜色分布和二进制分类器从包括相机传感器噪声的输入图像中分割前景区域和背景区域。 构成：基于二进制分类器的图像段单元（120）指定初始段图像中的区域，并通过指定区域和二进制分类器生成新的段图像。 图像段单元计算新分段图像的背景区域的直方图。 图像段单元通过时空图像块的加权和的和来更新背景模型生成器的时空图像块直方图。

公开(公告)号：KR1020100072845A

公开(公告)日：2010-07-01

申请号：KR1020080131372

申请日：2008-12-22

Applicant: 한국전자통신연구원

Inventor： 임성재 ,  김호원 ,  박일규 ,  박지영 ,  이지형 ,  김진서 ,  이승욱 ,  추창우 ,  황본우 ,  구본기

IPC: G06T13/40 ,  G06T17/10

CPC classification number: G06T17/10 ,  G06T5/30 ,  G06T13/40 ,  G06T17/30

Abstract: PURPOSE: A method for shape a deforming human body feature is provided to naturally supply automatic shape deformation of a feature point based Nurbs curve human body mode by automatically deforming a shape of a human body according to the motion capture data through a curve human body model. CONSTITUTION: A human body model generator(110) generates a key plane and feature point based human body mode data, and a statistical deformation information generator(120) creates statistical deformation information for a parameter of feature points, key plane and joint in specific motions. A deformation and animation unit(130) performs human body model shape deformation and animation based on the feature point based human body model and statistical deformation information.

Abstract translation: 目的：提供一种形状变形的人体特征的方法，以通过人体模型曲线自动地根据运动捕捉数据使人体形状自动变形，自动提供基于特征点的Nurbs曲线人体模型的自动形状变形 。 构成：人体模型生成器（110）生成键平面和基于特征点的人体模式数据，并且统计变形信息生成器（120）针对特定运动的特征点，关键平面和关节的参数创建统计变形信息 。 变形和动画单元（130）基于基于特征点的人体模型和统计变形信息进行人体模型形状变形和动画。

公开(公告)号：KR1020090110776A

公开(公告)日：2009-10-22

申请号：KR1020080125520

申请日：2008-12-10

Applicant: 한국전자통신연구원 ,  한양대학교 산학협력단

Inventor： 이승욱 ,  구본기 ,  김진서 ,  이지형 ,  김호원 ,  추창우 ,  황본우 ,  박정철 ,  박지영 ,  임성재 ,  박일규 ,  최윤석 ,  김갑기 ,  이영직 ,  장의선 ,  김대용 ,  김병준 ,  전재범 ,  손경수 ,  손기석

IPC: G06T17/00

CPC classification number: G06T9/001 ,  G06T17/20 ,  H03M7/3046 ,  H04N19/124 ,  H04N19/13 ,  H04N19/14

Abstract: PURPOSE: A low complexity three dimensional mesh compression device and a method thereof for encoding connection information by applying differential pulse code modulation are provided to improve compression rate and improve complexity of data compression about a three dimensional mesh model. CONSTITUTION: A low complexity three dimensional mesh compression device is as follows. A data analysis unit(510) separates geometry information(511), attribute information(512), and connection information(513). A mesh model quantization unit generates the quantized geometry information, the attribute information, and the connection information. A data modulation unit performs the differential pulse code modulation prediction by using he quantized value of the consecutive linkage information.

Abstract translation: 目的：提供一种低复杂度三维网格压缩设备及其通过应用差分脉码调制对连接信息进行编码的方法，以提高压缩率并提高关于三维网格模型的数据压缩的复杂性。 构成：低复杂度三维网格压缩装置如下。 数据分析单元（510）分离几何信息（511），属性信息（512）和连接信息（513）。 网格模型量化单元生成量化几何信息，属性信息和连接信息。 数据调制单元通过使用连续链接信息的量化值来执行差分脉冲编码调制预测。