National Cheng-Kung University

National Science Council

Other

United Microelectronics Corp.

Jar-Ferr Yang

Hung-Ming Wang

Hsi-Chun Tseng

Ke-Ying Liao

Chyou-Hsiung Hwang

Zhi-Hsien Chen

Kuan-Cheng Chen

Shih-Chang Hisa

Bor-Long Bai

Sheng-Yih Chen

Yung-Chiang Wei

Din-Yuen Chan

Chih-Peng Fan

A method comprises: obtaining two depth values from each of a first pixel depth value and a fourth pixel depth value, and obtaining one depth value from each of a second pixel depth value and a third pixel depth value; and assigning the two depth values obtained from the first pixel depth value to the R-subpixel and B-subpixel values of the first pixel, assigning the depth value obtained from the second pixel depth value to the R-subpixel, G-subpixel and B-subpixel values of the second pixel, assigning the depth value obtained from the third pixel depth value to the R-subpixel, G-subpixel and B-subpixel values of the third pixel, and assigning the two depth values obtained from the fourth pixel depth value to the G-subpixel value of the first pixel and the R-subpixel, G-subpixel and B-subpixel values of the fourth pixel.

Method and circuit of assigning selected depth values to RGB subpixels and recovering selected depth values from RGB subpixels for colored depth frame packing and depacking

An RGB format adjustment method includes: obtaining subpixel values having interleaved positions from four pixels in unadjusted RGB format; obtaining subpixel values of a first pixel in an adjusted RGB format according to the obtained subpixel values, wherein the R subpixel value of the adjusted first pixel is equal to an R subpixel value of the unadjusted first pixel, the G subpixel value of the adjusted first pixel is equal to an R subpixel value of a fourth pixel in the unadjusted RGB format, and the B subpixel value of the adjusted first pixel is equal to a B subpixel value of the unadjusted first pixel; and obtaining R subpixel values, G subpixel values and B subpixel values of a second pixel, a third pixel and the fourth pixel according to the obtained subpixel values and the obtained R, G and B subpixel values of the adjusted first pixel.

RGB format adjustment and reconstruction method and circuit for depth frame packing and depacking

A conversion method includes: obtaining two R's, two G's and two B's sub-pixel values having interleaved positions from four pixels of the RGB format; obtaining four Y-luminance value, one U-chrominance value and one V-chrominance value of the YUV format according to the obtained sub-pixel values. The Y-luminance values of the first, second, third and fourth pixels in the YUV format are calculated from R's of the first pixel, G's of the second pixel, G's of the third pixel and B's sub-pixel value of the fourth pixel respectively. The U-chrominance value of the first pixel is calculated from B's of the first pixel, R's of the first pixel and the G's sub-pixel value of the third pixel. The V-chrominance value of the first pixel is calculated from the R of the fourth pixel, G of the second pixel and B of the fourth pixel.

RGB to YUV format conversion and inverse conversion method and circuit for depth packing and depacking

A method, device and system for packing a color frame and an original depth frame to obtain a packed frame are disclosed. The color frame is corresponding to the original depth frame, and the packed frame is to be displayed on a screen. The method for packing the color frame and the original depth frame includes the steps of: resizing the original depth frame to obtain a resized depth frame; and combining the color frame and the resized depth frame to obtain the packed frame. The center of the color frame is displayed in the central area of the screen.

Method, device and system for packing color frame and original depth frame

An unpacking method, an unpacking device and an unpacking system of a packed frame are disclosed. The packed frame is to be displayed on a screen and includes a color frame and a resized depth frame, and the color frame is corresponding to the resized depth frame. The center of the color frame is displayed in the central area of the screen. The unpacking method includes the steps of: extracting the color frame and the resized depth frame from the packed frame; and restoring the resized depth frame to obtain an original depth frame.

Unpacking method, unpacking device and unpacking system of packed frame

A method, a device and a system for restoring a resized depth frame into an original depth frame are disclosed. The method for restoring a resized depth frame into an original depth frame includes the steps of: obtaining a first sub-pixel value from one pixel of the resized depth frame; storing the first sub-pixel value in all sub-pixels of a first pixel of the original depth frame; obtaining a second sub-pixel value of the pixel of the resized depth frame; and storing the second sub-pixel value to all sub-pixels of a second pixel of the original depth frame.

Method, device and system for restoring resized depth frame into original depth frame

A method, a device and a system for resizing an original depth frame into a resized depth frame. The method for resizing an original depth frame into a resized depth frame includes the steps of: obtaining two sub-pixel values from at least two pixels of the original depth frame, respectively; and storing the two sub-pixel values into corresponding two sub-pixels of one pixel of the resized depth frame, respectively.

Method, device and system for resizing original depth frame into resized depth frame

An adaptive scan method for image/video coding in accordance with the present invention comprises acts of calculating an average power of the transformed coefficients vector and acts rearranging the powers of the transformed coefficients with descending order in the data block according to the power of transformed coefficients; Therefore, the adaptive scan method is dependent on different prediction mode witch has been coded in H.264 standard, and provides better rate-distortion performance in entropy coding to the conventional zig-zag scan.

Adaptive scan method for image/video coding

An apparatus with new fixed-coefficient recursive structures for computing discrete cosine transforms with the power-of-two length is disclosed. The fixed-coefficient recursive structures are developed from exploration of periodicity embedded in transform bases, whose indices can form a complete residue system or a complete odd residue system. Distinctively, we found that properly selected fixed-coefficient filters achieve lower round-off errors than the nominal variable-coefficient ones for computing DCTs in finite-word-length machines.

Apparatus with selective fixed-coefficient filter for performing recursive discrete cosine transforms

A processor is provided for transforming N.times.N discrete cosine transform (DCT) coefficients F.sub.uv inputted from a run-length-code 5 (RLC) decoder and arranged in an input order into an image data f.sub.jk in an integrated circuit through a 2-D inverse discrete cosine transform (IDCT) procedure wherein subscripts u and v of DCT coefficients F.sub.uv are input frequency indices respectively having least significant bits (LSB) u.sub.0 and v.sub.0 having an exclusive-OR (XOR) and subscripts j and k of image data f.sub.jk are spatial indices generated by the integrated circuit, which comprises a cosine pre-multiplier array for computing cosine-weighted DCT coefficients, a principal subkernel mapper utilizing the cosine-weighted DCT coefficients by first referring to the indices u and v for forming a principal N/2.times.N/2 subkernel-weighted matrix F.sub.uv C.sub.1.sup.uv, an N.times.N accumulating matrix operating with the principal N/2.times.N/2 subkernel-weighted matrix F.sub.uv C.sub.1.sup.uv for progressively accumulating the image data f.sub.jk, and an output buffer for loading the image data f.sub.jk from the N.times.N accumulating matrix and transferring the image data f.sub.jk. Such a processor will reduce the number of the multipliers involved and simplify the hardware complexity therefor, fasten the pixel rate thereof so that the present method is especially suitable for the future HDTV systems, be able to progressively transform the arbitrary input order of DCT coefficients thereof so that the present method is especially suitable for the HDTV recording player systems and has an excellent regularity thereof and applies a simple hardware architecture so that the cost for the hardware is relatively low and the method is suitable for the manufacturing of VLSI for the HDTV systems.

Processor for performing two-dimensional inverse discrete cosine

In this invention, we introduce a connection indexing algorithm for realizing the arithmetic coding in finite wordlength processors. By determining the number of encoded information bits according to a difference between the finite wordlength and a residual termination parameter defined by the maximum difference of the information bits of any two adjacent symbols in the ordered-probabilities symbols set, the information of the last symbol is split into the current and the subsequent finite length codewords. The two or three encoded symbols split into the current and the subsequent finite length codewords are decoded independently, and the indices thereof are added to form a decoded symbol having an index equal to a sum of the addition.

Method and apparatus for finite-length arithmetic coding

A two-dimensional inverse discrete cosine transform (2-D IDCT) processor comprises cosine angle index generators, pipelined multipliers and a symmetrical kernel. The 2-D IDCT processor of the invention has a five-stage pipelined structure for carrying out a coefficient-by-coefficient 2-D IDCT algorithm and can be operated at a clock rate of more than 50 MHz to achieve a pixel rate of about 400 MHz.

Growing community of inventors

Tainan, Taiwan

Jar-Ferr Yang