Samsung Electronics Co., Ltd.

Sungkyunkwan University

Nyeong-Kyu Kwon

Woong-Hee Lee

Sung-Jei Kim

Yo-Won Jeong

Byeungwoo Jeon

Jong-Hwa Lee

Woonghee Lee

Kwanghyun Won

Yun-Seok Lee

Hyun-Ki Baik

Young-Beom Jung

Yowon Jeong

Jonghwa Lee

Hyuk-jae Jang

Jee-Hong Lee

Hideki Yamauchi

Se-Un Ryu

Kiran Varaganti

Kalyan K Kumar

Seungha Yang

Hyeon-Su Park

Insu Park

A video encoding device includes a pixel offset pre-calculation part configured to generate a loop-filtered pixel from a coding information and at least one of a source pixel, a predicted pixel and a restructured pixel, and configured to generate an offset from the loop-filtered pixel by performing a sample adaptive offset operation. The video encoding device also includes an entropy coding part configured to generate a bit stream from the offset.

Video encoding device and driving method thereof

An image processing device includes a bad pixel masking unit and a noise reduction unit. The bad pixel masking unit detects bad pixels in a plurality of blocks and masks the determined bad pixels. The noise reduction unit removes noise from the pixels in the plurality of blocks excluding the masked pixels using a pattern matching method and corrects bad pixels by obtained an average pixel value.

Image processing device and noise reduction system having the same

An image frame motion estimation device and image frame motion estimation method using the same include performing first sampling for generating a first sample source block by performing first-type sampling on pixels of a source block; performing second sampling for generating a first sample reference block by performing first-type sampling on pixels of a reference block; determining a first matching region by comparing pixel values of the first sample source block and the first sample reference block; and determining a second matching region corresponding to the source block by comparing pixel values of a plurality of regions adjacent to the first matching region and the source block. Where one pixel is sampled for each block constituted by N pixels (N is a natural number) in width and M pixels (M is a natural number) in height and the sampled pixels are projected in a horizontal direction, the first-type sampling causes at least two pixels to be sampled for every M pixels or where one pixel is sampled for each block constituted by N pixels (N is a natural number) in width and M pixels (M is a natural number) in height and the sampled pixels are projected in a vertical direction, the first-type sampling causes at least two pixels to be sampled for every N pixels.

Image frame motion estimation device and image frame motion estimation method using the same

A transform processor may process original data that includes video information by using at least one of first and second processing methods, and generates at least one of first and second transformed data. Index determination logic may determine an index satisfying a determination condition based on at least one of the first and second transformed data. Energy compaction determination logic may determine energy compaction of at least one of the first and second transformed data based on the determined index. Output selection logic may selectively output one of the first and second transformed data based on the determined energy compaction. An entropy encoder may encode data output from the output selection logic.

Video encoding circuit and video encoding method using the same, and operation processing device

An intra refresh method is provided. The intra refresh method includes dividing a first frame into regions, counting a number of intra macroblocks included in each of the regions, calculating weight values of the regions, configuring a macroblock included in a second region of a second frame as an intra macroblock, based on the number of the first region. The second region corresponds to the first region.

Intra refresh method for video encoding and a video encoder for performing the same

An image processing apparatus includes a region determiner configured to receive image data and perform a region determination by determining whether each of a plurality of sub-pixels included in the image data is included in an in-focusing region that is focused or an out-focusing region that is not focused; and an interpolator configured to perform demosaicing with respect to a sub-pixel included in the in-focusing region by using a first algorithm and perform the demosaicing with respect to a sub-pixel included in the out-focusing region by using a second algorithm, according to a result of the region determination, wherein, when the demosaicing with respect to the sub-pixel included in the in-focusing region is performed, one or more peripheral sub-pixels having phases that are different from a phase of the sub-pixel, on which the demosaicing is performed, are used.

Image processing apparatus, image processing method, and image processing system for performing sub-pixel interpolation

A multi core graphic processing device includes a first graphic core that processes a first segment of a graphic frame divided into a plurality of segments and generates a first local decision that defines a scene property of the first segment, a second graphic core that processes a second segment of the graphic frame different from the first segment and generates a second local decision that defines a scene property of the second segment, and a global decision unit that receives the first local decision and the second local decision from the first graphic core and the second graphic core, and selects one of the received first local decision and second local decision as a global decision.

Multi core graphic processing device

A video/image compression method based on an region of interest (ROI) can be performed in a variable block video encoder such as a HEVC. A ROI coding method using variable block size coding information partitions a maximum coding unit (LCU) block obtained from an image into coding unit (CU) blocks. To obtain a quantization parameter of each coding unit (CU) block, the value of its quantization parameter is assigned based on its first hierarchical depth information and its second hierarchical depth information by using the first hierarchical depth information related to the size of the coding unit (CU) block and using the second hierarchical depth information related to the size of a prediction unit (PU) block correspondingly represented in consequence of the partition of the coding unit (CU) block. The resulting values of a quantization parameter assigned to different coding unit (CU) blocks may be different.

Method and apparatus for ROI coding using variable block size coding information

To stabilize video (an image sequence), reconstructed block data and decoding information of a video frame are received by unit of macroblock from a decoding circuit. Global affine parameters are determined and provided based on the reconstructed block data and the decoding information, and the global affine parameters represent an affine transform of a frame. Stabilized block data are provided based on the global affine parameters by compensating the reconstructed block data for an affine motion corresponding to the affine transform.

Method of stabilizing video, post-processing circuit and video decoder including the same

A rate control method for multi-layered video coding, a video encoding apparatus and a video signal processing system employing the rate control method. In the rate control method for multi-layered video coding, encoding statistical information is generated based on the result of encoding input video data on a first layer. A second rate controller generates a plurality of quantization parameters to be used when encoding is performed on a second layer, based on the encoding statistical information and/or region of interest (ROI) information. Target numbers of bits that are to be respectively assigned to regions of a second layer are determined based on the encoding statistical information and/or ROI information, and the input video data is encoded at the second layer, based on the target numbers of bits.

Rate control method for multi-layered video coding, and video encoding apparatus and video signal processing apparatus using the rate control method

Provided are a depth information based optical distortion correction (ODC) circuit and method. The ODC circuit includes a depth acquisition unit acquiring a depth of an image, a grid generation unit dynamically generating a correction grid corresponding to the depth of the image using depth information of the image and a projection matrix, and a distortion correction unit correcting optical distortion in the image using the correction grid. The corrected image can then be stored, transmitted, or otherwise output. The image can be a single static image or photograph or a frame in a video recording.

Depth information based optical distortion correction circuit and method

A method and apparatus for encoding/decoding an image that divide an image sequence into sub-groups and determine encoding modes applied to bi-directional pictures included in each sub-group using correlations between the bi-directional pictures and reference pictures are provided. The image encoding method includes dividing a group of pictures (GOP) to be encoded according to consecutive B pictures into sub-groups; calculating the correlations between each B picture included in the sub-group and reference picture according to each encoding mode; and applying one of the encoding modes wherein the reference picture having the highest correlation with the B picture are used in each sub-group to encode the image, thereby improving encoding efficiency.

Growing community of inventors

Daejeon, South Korea

Nyeong-Kyu Kwon