Abstract:
PURPOSE: A multiview rendering device and a method thereof are provided to obtain the image information of adjacent view point without a performance by using a rendering result in a view point and perform rendering about a shielding area which does not obtain the image information. CONSTITUTION: A multiview rendering device renders at least one three-dimensional object based on a first view point(700). The device converts pixel values obtained by rendering into pixel values based on a second view point different from the first view point(720). The device detects a shielding area from an image based on the second view point(740). The device renders the detected shielding area based on the second view point(760). [Reference numerals] (700) Rendering at least one three-dimensional object based on a first view point; (720) Converting pixel values obtained by the rendering into pixel values based on a second view point different from the first view point; (740) Detecting a shielding area of the remaining area except for an area expressed by the pixel values, which are obtained by the conversion from an image based on the second view point; (760) Rendering the detected shielding area based on the second view point; (780) Generating the image based on the second view point by using the converted pixel values and the pixel values corresponding to the rendering result of the shielding area; (AA) Start; (BB) End
Abstract:
PURPOSE: A 3D mesh encoding device and a method thereof, and a 3D mesh decoding device and a method thereof are provided to simplify 3D meshes according to levels, thereby applying a progressive encoding method from a simple mesh of a low level to a complicated mesh of a high level. CONSTITUTION: An information determining unit(110) determines mesh information by a level. The mesh information includes connection information between vertexes and location information of each vertex. The vertexes form a 3D mesh. A bit stream generating unit(120) encodes the mesh information. The bit stream generating unit generates a bit stream.
Abstract:
PURPOSE: A KD tree creation method using a hardware and apparatus thereof are provided to support dynamically a video by creating a KD tree for a acceleration structure. CONSTITUTION: A cost calculation unit(310) reads triangle data from a memory by referencing the index of list data and searches the minimum cost location of a current node based on the triangle data. A left and right classification unit(320) classifies a right direction and a left direction based on the location. In case the cost calculation unit does not search the minimum cost location, a tree creation unit(330) creates an internal node based on a pop operation about a stack and creates a leaf node.
Abstract:
PURPOSE: A hybrid rendering device and a method thereof for selectively applying a rendering method according to the hardware performance are provided to utilize the advantage of each rendering mode and maximize rendering performance. CONSTITUTION: A determining unit(110) selects a rendering mode for performing a 3D rendering mode. A first rendering unit(120) expresses direct light according to the first rendering mode. A second rendering unit(130) performs the 3D rendering mode by expressing at least one indirect light according to a second rendering mode. A third rendering unit(140) performs a third rendering mode by expressing at least one of reflected light, refracted light and diffracted light.
Abstract:
PURPOSE: A device and a method for enhancing ray tracing speed are provided to perform packet tracing based on a group. CONSTITUTION: A first ray tracing performing unit(910) performs ray tracing about a primary ray. A secondary ray stores a secondary ray generated by corresponding to the ray tracing. A grouping performing unit(930) groups the secondary ray based on a direction of the secondary ray. A second ray tracing performing unit performs packet tracing about the secondary ray by the generated groups.
Abstract:
A rendering method and an apparatus thereof for displaying a text are provided to render the extracted text image by extracting a text image which a rendering device displays. A data converter(110) produces a character image showing character in consideration of text data. A rendering unit(120) renders the character image. The data conversion unit updates one of a location and a displayed size of the character. The text data expresses one of a character, a size, and a font. The data conversion unit inputs the two-dimensional video as the character image. The two-dimensional video is determined as the character image.