Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1998-11-17
2002-07-16
Rao, Andy (Department: 2713)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C375S240080
Reexamination Certificate
active
06421384
ABSTRACT:
TECHNICAL FIELD
This invention relates to a digital video compression technology and, more particularly, to novel systems and methods for determining contour-based motion estimation for compressing and transmitting video images so that they may be accurately reconstructed for providing quality images.
BACKGROUND ART
Since digital computers were introduced in the 1930's, they have been used in many areas of industry, including communications and in the video industry. One of the significant recent developments using digital computers involves data storage on an appropriate storage media and data communication (involving video data) transmitted through local area networks, and other communication network such as wide area network, Internet, World Wide Web, and others.
Video images as they can be seen on television or a computer screen are actually a series of still pictures. Each of the still pictures is called a frame. By showing the frames at a rapid rate, such as approximately 30 frames per second, human eyes can recognize the pictures as a moving scene. This invention concerns efficiently encoding and transmitting and accurately reconstructing and displaying video images.
For the purposes of this document, it will be useful to introduce terms with which the reader will need to be familiar in order to fully comprehend the disclosure contained herein. These terms are as follow:
B-frame: bidirectional predicted frame. A frame that is encoded with a reference to a past frame, a future frame or both.
Bitrate: the rate at which a device delivers a compressed bitstream to an input of another device.
I-frame: intra coded frame—a frame coded using information only from its own frame and not reference to any other frame.
I-VOP: intra coded video object plane—a video object plane coded using information only from the video object plane and not from any other video object plane.
IEC: International Electrotechnical Commission.
ISO: International Organization for Standardization.
Motion estimation: a process of estimating motion vectors for a video image.
MPEG: Moving Picture Experts Group. A group of representatives from major companies throughout the world working to standardize technologies involved in transmission of audio, video, and system data. Video coding standards are developed by the MPEG video group.
MPEG-1: a standard for storage and retrieval of moving pictures and associated audio on storage media. The current official denotation is ISO/IEC/JTC1/SC29/WG11.
MPEG-2: a standard for digital television at data rates below 10 Mbit/sec. The study began in 1990 and the standard for video was issued in early 1994.
MPEG-3: a standard initially to suit coding of high Definition TV (HDTV). MPEG-3 was later merged into MPEG-2.
MPEG-4: a standard for multimedia applications. This phase of standardization started in 1994 to accommodate the telecommunications, computer and TV/film industries.
MPEG-7: a content representation standard for various types of multimedia information.
P-frame: forward predictive frame. A frame that has been compressed by encoding the difference between the current frame and the past reference frame.
P-VOP: forward predictive video object plane. A video object plane that has been compressed by encoding the difference the video object plane and the past reference video object plane.
Pel: picture element in a digital sense. A pel is the digital version of a pixel in analog technology.
Video image: an image containing a video object, multiple video objects, a video object plane, an entire frame, or any other video data of interest.
VOP: video object plane as defined in MPEG-4. An image or video content of interest.
With the general meaning of this terminology in mind, a description of the general problems of the prior art and a detailed description of the operation of the invention are provided below.
Generally, when a video signal is digitized, a large amount of data is usually generated. For example, if a frame of a video image in a sequence of such frames is digitized as discrete grids or arrays with 360 pels (or pixels) per raster line and 288 lines/frame, approximately 311 Kbytes of memory capacity is necessary to store that one frame, assuming each pixel uses 8 bits of space to store color data. On a screen, a moving picture needs at least 30 frames per second to provide a realistic image. The raw data rate for a picture is about 72 Mbits per second or 4,320 Mbit (540 Mbyte) per minute. Therefore, it is almost impractical to store digital video data on a media or to send digital video data of several minutes to another location.
Moreover, real time transmission of video signals is impossible since no hardware currently available can provide the speed required to process the massive amount of data. Therefore, it is essential to compress the digital video data in order to generate moving pictures that are manageable using a current hardware technology.
A number of attempts have been made in the prior art to accomplish video data compression. Researchers discovered that the compression ratio of conventional lossless methods, such as Huffman, Arithmetic, and LZW, are not high enough for image and video compression. Fortunately, consecutive video pictures are usually quite similar from one to the next. Taking advantage of this, typically the prior art utilizes common video characteristics, such as spatial redundancy, temporal redundancy, uniform motion, spatial masking, and others to compress video picture data as used in Joint Photographic Expert Group (JPEG), H.261 compression, Moving Picture Experts Group (MPEG), and others.
One attempt to solve the problems of the prior art was made by a group called the Moving Picture Experts Group (MPEG) under the auspices of the International Organization for Standardization (ISO) and the International Electrotechnical Commission (IEC). Formed in 1988, this group has accomplished standardization of compression techniques for video, audio, and system which can be used throughout the world. Some of the standardization efforts of this group have been resulted in the standards known as MPEG-1 and MPEG-2 (into which MPEG-3 was merged). At the present time, the Group is working on an MPEG-4 standard.
A typical technique in compression, as adopted in the MEPG standard series, uses compression based on the Discrete Cosine Transform (DCT) and a motion compensation technique. The DCT-based compression is used to reduce spatial redundancy, and motion compensation is used to exploit temporal redundancy. Even though the Group working on MPEG-4 has adopted Shape Adoptive Discrete Cosine Transform (SADCT), the basic concept behind both DCT and SADCT is the same.
In MPEG-1 and MPEG-2, a frame can be usually encoded into three different types: intra-frame (I-frame), forward predictive frame (P-frame), and bi-directional predicted frame (B-frame). An I-frame is a frame that has been encoded independently as a single image without reference to other frames. A P-frame is a frame that has been compressed by encoding the difference between a frame and a past reference frame which is typically an I-frame or P-frame. A B-frame is a frame that has been encoded relative to a past reference frame, a future reference frame, or both. A typical group of encoded frames has a series of these types of frames in combination.
Each frame is typically divided into macroblocks. A macroblock consists of 16×16 sample array of luminance (grayscale) samples together with one 8×8 block sample for each of two chrominance (color) components. Macroblocks are the units of motion-compensated compression, and blocks are used for DCT compression.
When DCT compression is used, blocks are first transformed from the spatial domain into a frequency domain using the technique provided by DCT compression. Generally, DCT is a method of decomposing a block of data into a weighted sum of spatial frequencies. For example, an analog signal is sampled by discrete cosine functions with different spatial frequencies. Each of these spatial frequency patterns has a corresp
Chung Jae-Won
Kim Jae-Kyoon
Moon Joo-Hee
Park Cheol-Soo
Blakely & Sokoloff, Taylor & Zafman
Hyundai Curitel Inc.
Rao Andy
LandOfFree
System and method for contour-based motion estimation does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for contour-based motion estimation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for contour-based motion estimation will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2894655