Yonsei University

Colorzip Media, Inc.

Samsung Electronics Co., Ltd.

Colorzip Technology Co., Ltd.

Other

Hyundai Electronics Industries Co. Ltd.

Goldstar, Co., Ltd.

Intellectual Discovery Co., Ltd.

Tack Don Han

Woo Chan Park

Cheol Ho Cheong

Jae Yun Kim

Seong Hun Jeong

Jong Young Kim

Han Yeong Choi

Eui Jae Kim

Chan Min Park

Jeong Soo Park

Jae Ho Nah

Sang Yong Lee

Jin Woo Kim

Seok Yoon Jung

Woong Hee Jeong

Kil Whan Lee

Hyung Min Yoon

Gyeong Ja Jang

Keun Ho Kim

Jong Chul Jeong

Shin Jun Lee

Mo Young Suh

Il San Kim

Woo Nam Chung

Sung Bong Yang

Jae Won Ahn

Yun Hye Jung

Moon Key Lee

Nam Kyu Lee

Hyun Gu Shin

Chang Su Lee

Sang M Moh

Shi W Lee

An image processing apparatus is provided. A splitting unit of the image processing apparatus may split a first space within an input three-dimensional (3D) model into a plurality of subspaces in order to generate an acceleration structure of the input 3D model. A decision unit of the image processing apparatus may set a subspace determined as having a relatively high probability of including a ray progress path among the plurality of subspaces, as a child node having a relatively high traversal priority in the acceleration structure among a plurality of child nodes.

Ray tracing apparatus and method

An image rendering apparatus may include a buffer memory unit and a processor. The buffer memory unit may store input ray data for image rendering according to a ray tracing scheme while shape data corresponding to the input ray data is being fetched from a cache. The processor may output the received shape data together with the input ray data to an operation apparatus.

Image rendering apparatus and method for preventing pipeline stall using a buffer memory unit and a processor

Provided is an image processing apparatus. The image processing apparatus may perform an intersection test for rendering of a ray tracing scheme. The image processing apparatus may include a first calculator and a second calculator. The first calculator may perform a ray-plane test to determine whether a ray intersects a plane including a primitive and a barycentric test to determine whether the ray intersects the primitive. The second calculator may calculate a hit point based on the ray which intersects the primitive.

Image processing apparatus and method

A mixed code, and a method and apparatus for encoding the same are provided. The mixed code includes a first code image region storing a first code image obtained by encoding first information using color, shading, or a combination thereof, and a second code image region storing a second code image obtained by encoding second information using color, shading, or a combination thereof. The first code image and/or the second code image includes the results of encoding interpretation information, construction information, error control information, and code direction information. The mixed code is made by adjusting the difference in color and brightness between the first and second code images to a predetermined level and combining these images.

Mixed code, and method and apparatus for generating the same

Provided are a method and apparatus for decoding a mixed code that is a combination of first and second code images. In the method, a mixed code image is obtained by receiving the original image having the mixed code image and removing noise from the original image. Next, the colors, shades, and brightnesses of pixels of the mixed code image are categorized into groups based on a predetermined threshold, the mixed code image is divided into first and second code images, and first and second information are extracted by decoding the first and second code images, respectively. The first and second code images can be effectively decoded by decoding interpretation information, construction information, error control information, and code direction information stored in the first and/or second code images.

Method and apparatus for decoding mixed code

Provided is a method of classifying colors of a color based image code. The colors of each cell forming the color based image code are classified by sampling a predetermined number of pixels from each cell and then applying a max channel based clustering method, a white balance based color clustering method, a single-linkage algorithm based color classification method, a K-means algorithm based color classification method, or the like. Accordingly, the colors of an image code can be accurately recognized despite of a color distortion due to characteristics of a camera, a printing medium, etc.

Method of classifying colors of color based image code

A method, medium, and system of rendering image data, including receiving three-dimensional space data for each pixel forming the image data, the three-dimensional space data including depth and color information values for each pixel, determining an order of rendering for a plurality of surfaces forming the image data in an order of the distance between a rendering viewpoint and the surface forming the image data, generating splat depth information values for each splat included in the surface forming the image data, according to the rendering order, and determining whether the splat included in the surface forming the image data is required to be blended and accordingly selectively performing a blending on the splat.

Depth image-based rendering method, medium, and system using splats

A method and device for generating and decoding an image code without a direction finder pattern are provided. In the method of generating an image code, a two-dimensional image is generated by encoding the configured basic information and the error correction information, and disposing the encoded basic and error correction information in predetermined regions, respectively. It is determined whether the two-dimensional image can be decoded for directions which are different from a basic direction of the two-dimensional image, respectively, and the two-dimensional image is output as a physical or electronic image code when a result from decoding for the directions which are different from the basic direction of the two-dimensional image is the same as a decoding result for the basic direction, or when decoding cannot be performed for a direction which is different from the basic direction. In the method of decoding an image code, a physically or electronically represented image code is received as an input, and a decoding result for a direction for which the decoding result has the least number of errors from among the results decoded for each direction is output. Accordingly, a direction of an image code can be detected without any overhead or with a minimal overhead.

Method and device for generating image code and method and device for decoding image code

A hybrid tag interface system and an interfacing method therefor are provided. First codes for an object from a wireless recognition tag containing the first codes are received through wireless communication, a second code for the object from an image tag expressing the second code by using a 1- or 2-dimensional image is read through optical contact, and then information on the object is calculated from the first codes and the second code based on their correlation. By doing so, automatic recognition and visibility can be maximized using the hybrid tag combining a wireless recognition tag and an image tag.

Hybrid tag interface system and method using the hybrid tag interface

A pipelined divider with a small lookup table is disclosed. The pipelined divider can greatly reduce the size of a lookup table with a low cost to overcome the problems involved in the conventional pipelined divider requiring a large lookup table due to its iterative operation type. The pipelined divider has a delay time of 3 cycles in a single precision, and can reduce a chip size by about &#x2153; in comparison to the existing pipelined divider.

Pipelined divider and dividing method with small lookup table

An apparatus and method for recognizing a code from a code image that is expressed physically or electronically and extracting data represented in the code image is provided. The method includes the steps of receiving a raw image in which a code image is contained, detecting a background image included in the raw image, extracting a code image region in which a background image is excluded, recognizing the shape and type of the code image and the color or shade represented in each of cells, converting the color or shade recognized from each of the cells into a corresponding character, number, or symbol and generating code data. The code image in which predetermined data are represented as colors or shades are received, and original colors or shades can be precisely discriminated regardless of an environment in which the code image is recognized.

Apparatus and method for recognizing code

A method and a device of consistency buffer for a high performance 3D graphic accelerator is disclosed to retain consistency without detecting any overlapping region in advance but determining an overlapping with respect to a rendered pixel. The device of consistency buffer according to the present invention comprises a fetch section for transmitting a plurality of primitives to be processed to a vacant region of the buffer, an issue section for buffering the plurality of the primitives transmitted from the said fetch section, and allotting positions to be inputted, a plurality of rendering accelerators for receiving and rendering the plurality of primitives allotted by the issue section, a consistency buffer for storing information required for processing with consistency according to a depth value and a color value of each primitive computed by each of the rendering accelerators, and a memory interface unit for performing read/write computation in a memory by mens of processing with consistency in order based on the information stored in the consistency buffer.

Method and device of consistency buffer for high performance 3D graphic accelerator

A method and apparatus for processing pixel rasterization in a 3D rendering processor is disclosed. According to the method and apparatus, the primary depth checking is performed before the performing of the texture mapping, and thus the unnecessary performing of the texture mapping can be removed. Also, the consistency problem can be simply and easily solved using the flag memory, and by performing the depth reading and depth checking twice, the hit rate of the pixel cache memory is heightened. Thus, the method and apparatus is effective in cost, performance, and power consumption.

Method and apparatus for processing pixel rasterization in three-dimensional rendering processor

Growing community of inventors

Seoul, South Korea

Tack Don Han