Deutsche Thomson-brandt Gmbh

Thomson Licensing

Telefunken Fernseh Und Rundfunk Gmbh

Deutsche Ag Fur Unterhaltungselectronic Ohg

Duetsche Thomson-brandt Gmbh

Heinz-Werner Keesen

Hartmut Peters

Carsten Herpel

Wolfgang Hartnack

Marco Winter

Harald Schiller

Hans-Joachim Platte

Herbert Schutze

Gunter Oberjatzas

Ralf Ostermann

Ernst F Schroder

Dietmar Hepper

Fritz Weisser

Dietmar Uhde

Jens Spille

Jurgen Kaaden

Michael Drexler

Jens Peter Wittenburg

Martin Plantholt

Laurent Perdrieau

Herbert Schütze

Rolf Schiering

Dieter Haupt

Timothy Heighway

Hans-Hermann Hake

Siegfried Schweidler

Motion compensated temporal prediction is well-known for redundancy reduction in video signal compression. A prediction error signal and a motion vector for a current block are transmitted, describing the location of the predictor block in the reference picture. A problem solved by the invention is to encode and decode video signals using less or no mo- tion vector data side information. This is achieved through a temporal prediction which exploits common knowledge available in both encoder and decoder, requiring that the decoder performs a motion estimation, too. A common baseline motion estimation process is used at both encoder and decoder on the same set of video data that have already been compressed before and, hence, are available at both encoder and decoder, leading to identical motion parameters being generated in the decoder. These common motion parameters are used for generating a default temporal prediction without the need to transmit related motion vectors from the encoder to the decoder.

Method and apparatus for generating coded picture data and for decoding coded picture data

In MPEG, most video frames are encoded in inter mode. Every now and than a video frame is encoded in intra mode in order to limit error propagation and to enable access points for starting decoding of the video sequence. According to the invention, however, it is sufficient to encode in an intra frame only those pixel blocks in intra mode of which at least one pixel is used for predicting some future picture content information, whereas the other blocks can be encoded in inter mode. In order to check which blocks in the intra frame can be encoded in inter mode, a motion compensation pre-analysis of the following P-frame is performed, in order to find those blocks that need not be encoded in intra mode because no pixel in such blocks is used for the prediction of that following P-picture.

Method and apparatus for encoding a picture sequence using predicted and non-predicted pictures which each include multiple macroblocks

In bitstream recording presentation data is organized into Video Object Units. These have a variable size but have also a variable duration. To allow access to any Video Object Unit in the bitstream a housekeeping address table is used which is based on pieces of the bitstream of constant size per piece. The address table additionally contains for each of these pieces a specific delta duration which indicates the time difference between the arrival time of the first packet of a piece and the arrival time of the packet following immediately the last packet of that piece. The computation of the target VOBU address includes the following steps:

Method for addressing a bitstream recording

Today the storage of audiovisual (AV) data on storage media employs standardized coding schemes, e.g. MPEG-2. End user devices for AV presentation, like disc players, contain dedicated decoders implemented in hardware, which are able to decode data streams complying with the respective standard. The efficiency of this decoding methodology relies on the standardization of appropriate coding schemes. Proprietary coding schemes may be more efficient by being adapted to the content, or may be used for content protection or optional features. Using reconfigurable decoder hardware, storing the configuration data for the actual decoder on the same storage medium as the AV data, and downloading the data to configure the player can be used to implement proprietary coding schemes. Such schemes require appropriate player hardware, suitable to execute a downloaded decoder under real-time conditions, and a standardized format to store the decoder configuration data.

Method for decoding data received from a data source using hardware configuration data received from the same data source

In bitstream recording presentation data is organized into Video Object Units. These have a variable size but have also a variable duration. To allow access to any Video Object Unit in the bitstream a mapping list is used which is based on pieces of the bitstream of constant size per piece. The mapping list additionally contains for each of these pieces a specific delta duration which indicates the difference between the arrival time of the first packet of a piece and the arrival time of the packet following immediately the last packet of that piece, and contains an initialization value which allows to have a share of equal bits between a predetermined amount of MSBs of application time stamps and the same amount of LSBs of the summation of said delta time durations. The computation of the target SOBU address includes accumulating the initialization value and the delta durations until the given time value is most closely reached towards the target SOBU then multiplying the running index of this table entry by the constant piece size directly resulting in the address value to be accessed.

A set top box can be connected to a DVD Streamer via an IEEE1394 interface which contains means to timestamp data and to strip off these timestamps again, using them for timing regeneration. The DVD stream also must regenerate the timing of data packets as it was upon recording, when these packets are played back. The streamer could also use its own timestamps and strip them off again when replaying. So, in total, there is an in-series connection between two time stamping and time regeneration mechanism which introduces jitter accumulation.

Method and apparatus for timestamping a bitstream to be recorded

The IEEE 1394 bus communication protocol has three layers: physical layer, link layer, and transaction layer. Typically, the transaction layer is realized by firmware whereas the other layers are implemented using chip sets. The link layer IC usually contains a FIFO having a capacity of e.g. 32k or 64k bits. Therefore, the link layer chip is the most costly part of a complete IEEE 1394 interface. Due to these cost reasons most ICs on the market are not bi-directional although the IEEE 1394 bus specification supports this feature. Incoming or outgoing data packets are intermediately stored in the FIFO. The current solution to this problem is to have two separate IEEE 1394 bus nodes assigned to the same application, the two nodes including two physical layer ICs and two link layer ICs. The inventors have found that although the physical link layer interface is not designed for this purpose, it works correctly with up to three link layer ICs and one physical layer IC if the additional link layer IC/ICs is/are programmed respectively. Therefore two or more link layer ICs can operate together with one physical layer IC in one node wherein the link layer ICs are connected to the same application or device.

Method and apparatus for bi-directionally transferring data between an IEEE 1394 bus and a device to be controlled by a control signal transmitted via the bus

The computation-intensive task of motion estimation for video signals is performed by block search methods using apparatus of reduced complexity. Motion estimation is performed by analyzing only specific bits, e.g. one or two MSBs, of the picture element values. These bits are logic-linked by means of EXOR and/or EXNOR functions, and the individual results for different block positions of a current block within a search window are combined. The respective motion vector for the current block is determined by using a minimum or maximum comparison.

Video encoder and decoder image motion estimation system with reduced

A digital video recording apparatus comprises a means for receiving a coded video image representative signal comprising components representing spatial frequencies including DC and AC frequency information. A grouping means groups components of substantially the same type to produce grouped signals for recording. A recording means records the grouped signals on a recording medium.

Digital video signal processing for recording and replay

A cassette for use with a recording apparatus comprises: a housing for a recording medium, the housing having a first surface and side walls, the recording medium being accessible through one of the side walls; at least a part of a chip card affixed to the housing, forming a second surface of the housing and closing the cassette, the chip card part having a memory formed integrally therewith and externally accessible contacts; and, the contacts of the chip card being positioned on the second surface to be engageable by corresponding contacts in the recording apparatus when the cassette is inserted therein. One of the side walls has an access door for the recording medium and the second surface forms an upper wall of the housing as defined by the access door. The part of the chip card forming the second surface of the housing has a marking position at which light transmissivity can be controlled. The marking position can be rendered non transmissive to light to prevent recording.

Cassette including a recording medium for a recording apparatus

An apparatus for the transmission and reception of block coded discrete cosine transformed digital video signal having coefficients representing a DC component and a plurality of AC components of various importance. A means responsive to the digital signal separates and stores like components of the transformed digital signal. The like components are assembled into segments which define a portion of a picture. Addresses of the segments are generated and added thereto. The segments are modulated for transmission by a modulating means. Upon reception the modulated segments are demodulated by a demodulating means. The like components of segments are stored responsive to the addresses. A means, coupled to the storing means and responsive to the addresses, reconstructs from the like components, the block coded discrete cosine transformed digital video signal.

Digital video signal processing system for transmission and recording

A method of making a hierarchical estimate of image motion for a television signal. A television signal video image is divided into blocks of fixed size, and motion relative to a second video image is determined for each block. Mutually non-overlapping pixel blocks are produced at a lower, more finely resolved hierarchical level. The center points of these blocks are used as the points for an analysis at a level that is located higher in the hierarchy. Preferably, points which lie between image lines and which do not represent real picture elements in the line raster, are investigated in the mutually non-overlapping blocks.

Method of making a hierarchical estimate of image motion in a television

Customary recorders record data packets of which the format is adapted to the data format of the recorder. If, however, the data packets have a differing format, various difficulties arise, for example, with error protection. In such cases, therefore, there is also no provision for a search mode with an at least partial evaluation of data packets. According to the invention, before the recording/storage, the data packets are consequently arranged 'transversely' in accordance with FIG. 1, i.e. it is not so that each data packet (1, 2, 3, 4, 5) is progressively recorded/stored (in the vertical direction in FIG. 1), but that the first section (a) of the data packets is progressively recorded (in the horizontal direction in FIG. 1). Subsequently, the further sections of these data packets are recorded. In the fast search of a digital recorder, the helical tracks can, however, only be read section by section. Of the data packets (1 . . . 5), there would only be picked up rudiments, which cannot be meaningfully decoded for reproduction. Thus, additional data packets (A1, A2, B1, B2), which may also partially be copies of the data already recorded in the 'normal' way, are recorded in reserve sections which can be contiguously read in a search. In a search at increased speed, these reserve sync blocks are then evaluated.

Method of storing or reproducing data packets

A hybrid Discrete Cosine Transform (DCT) coder processes blocks of (e.g., 8.times.8) pixels from interlaced or progressive scanned image signals. Processing of image lines by the coder is modified in the presence of image motion to avoid switching between 8.times.8 and 2.times.(4.times.8) transformations. For motion, the lines of two vertically superimposed image blocks are rearranged to produce first and second modified blocks. The modified first block contains pixels from lines in a first field, and the modified second block contains pixels from lines in an associated adjacent second field.

System for coding an image representative signal

In a system for generating a compatible television signal with a 'letterbox' display format, groups of vertically arranged picture elements (pels) are subjected to vertical DCT transformation producing eight DCT coefficients. Intermediate sixth and seventh coefficients are placed in the letterbox bar regions. The remaining six coefficients, namely the first through fifth coefficients and the highest frequency eighth coefficient, are inverse DCT transformed and processed with the main image information. A standard aspect ratio receiver displays a letterbox image using only the remaining six coefficients. A wide aspect ratio receiver uses the six remaining coefficients with the two coefficients from the bar regions, after inverse DCT, to produce a full resolution wide aspect ratio image.

Growing community of inventors

Hannover, Germany

Heinz-Werner Keesen