-
公开(公告)号：KR101788583B1
公开(公告)日：2017-10-23
申请号：KR1020110011025
申请日：2011-02-08
Applicant: 삼성전자주식회사 , 서강대학교산학협력단
Abstract: 복셀의가시성에기반하여 kd-트리를구축하기위한방법및 장치가제공된다. 복셀의가시성은씬 내에서계속이동할레이들에대해상기의복셀이얼마나가시적인지를나타낸다. 복셀의가시성에기반한비용함수가 kd-트리를구성하기위해사용된다. 비용함수를최소화하는평면을사용하여복셀을 2 개의부-복셀들로분할함으로써 kd-트리가구성된다. 구성된 kd-트리는레이트레이싱을수행하기위해사용된다.
-
公开(公告)号：KR1020120090543A
公开(公告)日：2012-08-17
申请号：KR1020110011025
申请日：2011-02-08
Applicant: 삼성전자주식회사 , 서강대학교산학협력단
CPC classification number: G06T15/06 , G06T15/08 , G06T17/005 , G06T2200/04
Abstract: PURPOSE: A voxel visibility based KD(K Dimension) tree configuration apparatus and method thereof are provided to estimate the probability of a node tour using a voxel based matrix. CONSTITUTION: A voxel visibility calculation unit(810) calculates voxel visibility. A KD tree configuration unit(820) divides the voxel into a first voxel unit and a second voxel unit using a surface which minimizes voxel visibility based cost functions. The KD tree configuration unit configures the KD tree by dividing the voxel. A ray tracing unit(830) traces ray for a scene using the configured KD tree.
Abstract translation: 目的：提供一种基于体素可视性的KD（K Dimension）树配置装置及其方法，以估计使用基于体素的矩阵进行节点巡视的概率。 构成：体素可视性计算单元（810）计算体素可视性。 KD树配置单元（820）使用使得基于体素可视性的成本函数最小化的表面将体素划分为第一体素单元和第二体素单元。 KD树配置单元通过划分体素来配置KD树。 光线跟踪单元（830）使用配置的KD树跟踪用于场景的射线。
-
公开(公告)号：KR1020120070824A
公开(公告)日：2012-07-02
申请号：KR1020100132304
申请日：2010-12-22
Applicant: 서강대학교산학협력단
CPC classification number: G06T17/30 , G06T15/06 , G06T17/005
Abstract: PURPOSE: A method for constructing a data structure is provided to update a data structure of a previous frame and to construct a data structure based on a polygon significance. CONSTITUTION: Polygon information is extracted(S100). An initial cross test acceleration data structure is setup(S110). A ray is traced for a first frame(S120). The cross test acceleration data structure is generated by deformation of the initial cross test acceleration data structure(S140). A ray is traced about the next frame(S150).
Abstract translation: 目的：提供一种用于构建数据结构的方法，以更新前一帧的数据结构，并构建基于多边形有效性的数据结构。 构成：提取多边形信息（S100）。 设置初始交叉测试加速度数据结构（S110）。 跟踪第一帧的光线（S120）。 交叉测试加速度数据结构由初始交叉加速度数据结构的变形产生（S140）。 关于下一帧跟踪射线（S150）。
-
-
公开(公告)号：KR101956312B1
公开(公告)日：2019-03-08
申请号：KR1020170022963
申请日：2017-02-21
Applicant: 서강대학교산학협력단
IPC: G06T7/70
-
Patent Agency Ranking
公开(公告)号：KR101653203B1
公开(公告)日：2016-09-02
申请号：KR1020100009025
申请日：2010-02-01
Applicant: 삼성전자주식회사 , 서강대학교산학협력단
CPC classification number: G06T15/50
Abstract: 영상처리장치가제공된다. 렌더링하고자하는영상의픽셀중 복수개의참조픽셀에대해통상적인방법에의해레이트레이싱이수행된다. 그리고, 상기참조픽셀이아닌픽셀에대해서는주변의참조픽셀의레이트레이싱결과를참조하여기하정보가보간될수 있다.
-
公开(公告)号：KR1020140070869A
公开(公告)日：2014-06-11
申请号：KR1020120136326
申请日：2012-11-28
Applicant: 서강대학교산학협력단
CPC classification number: G06F3/0416 , G01C21/00 , G06F3/0412 , G06F3/0484 , G06F3/0488 , H04N5/225
Abstract: A rotation conversion method according to the present invention includes the steps of: (a) setting the vertical axis of an initial camera frame as a fixed rotation shaft; (b) converting the rotation by using an Euler angle while using the fixed rotation shaft of the initially set camera frame when a rotation operation is input as the user continuously touches a touchscreen; (c) resetting the fixed rotation shaft as the vertical axis of the camera frame at the point of the end of the touch when the touch on the touchscreen is completed; (d) repeatedly performing the step (b) while using the rotation axis of the reset camera frame axis as a reference shaft.
Abstract translation: 根据本发明的旋转转换方法包括以下步骤：（a）将初始相机框架的垂直轴设置为固定旋转轴; （b）当用户连续触摸触摸屏时输入旋转操作时，使用初始设定的相机框架的固定旋转轴时，通过使用欧拉角度来转换旋转; （c）当触摸屏上的触摸完成时，在触摸结束点的位置将固定旋转轴重新设置为相机框架的垂直轴线; （d）在使用复位摄像机框架轴的旋转轴作为基准轴的同时重复执行步骤（b）。
-
公开(公告)号：KR101373174B1
公开(公告)日：2014-03-11
申请号：KR1020120120625
申请日：2012-10-29
Applicant: 서강대학교산학협력단
IPC: G06T15/06
CPC classification number: G06T15/06
Abstract: A method for processing polygon data of a spatial acceleration structure according to the present invention includes a step of detecting overlappingly stored polygon data from inner nodes included in a spatial acceleration structure by leading the polygon data stored in leaf nodes included in the inner node; and a step of selecting the polygon data overlappingly stored in the leaf nodes as reconstitution target polygon data, storing the same in the inner node, and removing the polygon data in the leaf nodes. The polygon data is data on a polygon identification number. [Reference numerals] (10) Processing unit; (100) Memory unit; (102) Existing spatial acceleration structure; (104) Reconstituted spatial acceleration structure; (112) Gradual spatial acceleration structure reconstructing unit; (114) Data layout change unit on a polygon identification number
Abstract translation: 根据本发明的用于处理空间加速结构的多边形数据的方法包括通过引导存储在包括在内部节点中的叶节点中的多边形数据来检测包括在空间加速度结构中的内部节点的重叠存储的多边形数据的步骤; 并且将重叠存储在叶节点中的多边形数据作为重构目标多边形数据，将其存储在内部节点中，以及去除叶节点中的多边形数据的步骤。 多边形数据是多边形标识号的数据。 （附图标记）（10）处理单元; （100）存储单元; （102）现有空间加速度结构; （104）重构空间加速度结构; （112）渐进空间加速度结构重建单元; （114）多边形识别号码上的数据布局变更单位
-
公开(公告)号：KR1020100118706A
公开(公告)日：2010-11-08
申请号：KR1020090037527
申请日：2009-04-29
Applicant: 삼성전자주식회사 , 서강대학교산학협력단
CPC classification number: G09G5/14 , G06T15/503 , G06T2200/12 , G09G2320/02 , G09G2340/0457 , G09G2340/10
Abstract: PURPOSE: An image processing apparatus for anti-aliasing and a method thereof are provided to reduce unnecessary operation quantity by selecting a sub block necessary for super sampling. CONSTITUTION: An image processing apparatus(100) comprises a selecting unit(110), a processing unit(120) and a buffer memory(140). The selecting unit selects one sub pixel for anti-aliasing in a first pixel. The processing unit processes the anti-aliasing in each of selected one or more sub pixel and the color value of the first pixel. The buffer memory stores sub pixels which the selecting unit selects.
Abstract translation: 目的：提供一种用于抗锯齿的图像处理装置及其方法，通过选择超采样所需的子块来减少不必要的操作量。 构成：图像处理装置（100）包括选择单元（110），处理单元（120）和缓冲存储器（140）。 选择单元在第一像素中选择一个子像素用于抗锯齿。 处理单元处理所选择的一个或多个子像素中的每一个中的抗混叠和第一像素的颜色值。 缓冲存储器存储选择单元选择的子像素。
-
-
-
-
-
-
-
-
-
Search Results
21
records
(took 0.021 seconds)
