Wei C Ding

US Patent:

6850566, Feb 1, 2005

Filed:

Feb 20, 2001

Appl. No.:

09/937058

Inventors:

Chung-Tao Chu - San Jose CA, US
Wei Ding - Fremont CA, US

Assignee:

InterVideo, Inc. - Fremont CA

International Classification:

H04N 712

US Classification:

37524003

Abstract:

A system for improving the speed of the video encoding process by decreasing the number of cycles to perform the quantization. The disclosed system achieves the improvement through use of parallel processor such as one having a Single Instruction, Multiple Data (SIMD) architecture. Concurrent processing during one instruction cycle is accomplished, thereby leading overall to the use of fewer instruction cycles. In the preferred embodiment, an MMX instruction set is used for executing four quantization in parallel. The disclosed system also achieves a higher precision of the quantization during the encoding of video signals with the SIMD architecture by using a larger multiplier and larger shift factor.

US Patent:

6972801, Dec 6, 2005

Filed:

Oct 20, 2000

Appl. No.:

09/693351

Inventors:

Wei Ding - Fremont CA, US
Andy C. Hung - Los Altos CA, US

Assignee:

Intervideo, Inc. - Fremont CA

International Classification:

H04N005/21

US Classification:

348448

Abstract:

A system and method for processing a video signal to make the corresponding image suitable for display on a progressive display device, i. e. , a computer monitor, is disclosed. The processing system includes a video display module which is capable of processing an image frame based on the content of the frame information itself, or the values of the pixels that comprise the frame.

US Patent:

58838236, Mar 16, 1999

Filed:

Jan 15, 1997

Appl. No.:

8/784739

Inventors:

Wei Ding - Fremont CA

Assignee:

Sun Microsystems, Inc. - Mountain View CA

International Classification:

G06F 1714

US Classification:

36472503

Abstract:

An array of DCT transform coefficients are converted to a two-dimensional array of spatial data in a video compression or decompression system. The array of DCT transform coefficients are divided into two groups. A regional IDCT algorithm is applied to all coefficients, both non-zero and zero, of the first group, while IDCT computation is applied to only non-zero coefficients of the second group. The results of the operations are then combined and/or mapped to form the output array of spatial data. In one specific implementation wherein an 8. times. 8 array of DCT coefficients are to be transformed, the first region is defined by the first quadrant of coefficients in the 8. times. 8 array. A regional IDCT algorithm is applied to both zero and non-zero coefficients in the first quadrant. The non-zero coefficients in the remaining three quadrants are calculated directly, and the results of the direct calculations are combined with the results of the regional IDCT algorithm applied to the first quadrant coefficients.

US Patent:

62254510, May 1, 2001

Filed:

Mar 4, 1999

Appl. No.:

9/262773

Inventors:

Dennis G. Ballinger - Menlo Park CA
Wei Ding - Salt Lake City UT
Susanne Wagner - Murray UT
Mark A. Hess - Salt Lake City UT

Assignee:

Myriad Genetics, Inc. - Salk Lake City UT

International Classification:

C07H 2104
C12Q 168

US Classification:

536 221

Abstract:

Human coronary heart disease susceptibility gene (CHD1), some alleles of which are related to susceptibility to coronary heart disease. Germline mutations in the CHD1 gene and their use in the diagnosis of predisposition to coronary heart disease and to metabolic disorders, including hypoalphalipoproteinemia, familial combined hyperlipidemia, insulin resistant syndrome X or multiple metabolic disorder, obesity, diabetes and dyslipidemic hypertension. Presymptomatic therapy of individuals who carry deleterious alleles of the CHD1 gene (including gene therapy, protein replacement therapy, and administration of protein mimetics and inhibitors). The screening of drugs for dyslipidemic therapy.

US Patent:

60056266, Dec 21, 1999

Filed:

Jan 9, 1997

Appl. No.:

8/780780

Inventors:

Wei Ding - Fremont CA

Assignee:

Sun Microsystems, Inc. - Palo Alto CA

International Classification:

H04N 712
H04N 1102
H04N 1104
H04N 521

US Classification:

348416

Abstract:

Motion vectors and associated errors which are determined in the course of encoding a video signal are used to determine whether respective macroblocks of two or more frames represent the same subject matter albeit displaced by motion represented by the frames. The motion vector errors are compared to a predetermined motion threshold. If a particular motion vector error is less than the motion threshold, the macroblocks corresponding to the motion vector error are determined to represent substantially identical subject matter and the corresponding motion vector represents displacement of the subject matter between the frames due to motion of the subject matter. Accordingly, differences between the macroblocks corresponding to the motion vector error are determined to be due primarily to noise and not to differences in the subject. Therefore, a relatively heavy temporal filter is applied between the macroblocks. Conversely, if a particular motion vector error is not less than the motion threshold, the macroblocks corresponding to the motion vector error are determined to represent different subject matter and only a spatial filter is applied to either macroblock.