Abstract: A method and apparatus for decompressing data in a data-compression system with decoder-only side information is provided. In one aspect, the method comprises generating side information using a source reconstruction and decoding using the generated side information to generate a new source reconstruction. The method further includes iterating the steps of generating and decoding, the generating step using at least the new source reconstruction output by the previous decoding step, and the decoding step using the side information output by the previous generating step. The method may stop the iteration when one or more predetermined criteria are met.

Abstract: A computer-based method for encoding and decoding quantized sequences in Wyner-Ziv coding of video. The method takes as input to an encoder a sequence of quantized data from a finite alphabet, the sequence comprising a video frame. Another sequence of data, decoded from the prior video frame, is input to a decoder. A statistical model describing the statistical relationship between the quantized input sequence to the encoder and the input sequence to the decoder is input to both the encoder and the decoder. A minimum field size is estimated from the statistical model at the encoder and the decoder separately. A bi-partite graph is provided, having variable nodes and check nodes, in which the alphabet of each check node is the minimum field size. At the output of the check nodes of the bi-partite graph there is generated a syndrome sequence of data in response to the quantized sequence fed into the variable nodes.

Abstract: The present invention relates to computer-based systems and methods for visual signal extrapolation or interpolation based on statistic similarity estimation. Specifically, a first and a second reference pictures are provided, and motion estimation is conducted on the first and second reference pictures to generate motion vectors indicative of movement of at least one of the first and second reference pictures in relation to the other. Subsequently, an estimate picture is generated by extrapolation or interpolation from the first and/or the second reference picture using the motion vectors, followed by optional refinement of the estimate picture. Statistic similarity estimation is used either in motion estimation or in refining the estimate picture, or a combination of both, so as to provide improved visual signals.

Abstract: A novel video surveillance system is made up of video and audio compression engine, a storage device and, a video and audio recognition engine. The video recognition engine detects such events as face recognition, motion detection etc, whereas audio recognition engine detects voice and other sound signatures indicating a potential alarm situation, e.g., panic voices such as screaming and yelling, or sounds such as gun shots, explosions. Combined recognition of audio and video signals provides for higher true alarm generation and lower false alarms level of the surveillance system. Additionally, the audio recognition engine provides information for directing video cameras in the direction of interest allowing better capture of an interesting scene.

Abstract: A method and apparatus is disclosed for presenting information to a user of an audiovisual appliance such as a television receiver, a set top box, or a DVD or a VCR player. The method includes extracting source text from a graphical representation of the source text in an associated audiovisual program; translating the extracted source text into a target text in a language different than a language of the source text; and presenting the target text in a human perceptible format in a synchronized manner with the associated audiovisual program. Translating the extracted source text may include an initial step of identifying the source language.

Abstract: A system and method of on-line handwriting recognition includes screen pen data to be combined with other information obtained through the usage of built-in digital camera, such as pen tilt parameter information. The method and system provide a low-cost solution for achieving better handwriting recognition quality for consumer platforms such as PDA. Camera being built-in does not increase the cost of the solution—the only added component is software-based pattern recognition algorithm that extracts non screen-related pen information.

Abstract: A real-time MPEG video coding system with information look-ahead for constant bit rate (CBR) applications, such as, for example, Video-on-Demand (VoD) over ADSL. This scheme employs two MPEG encoders. The second encoder has a buffer to delay the input by an amount of time relative to the first encoder to create a look-ahead window. In encoding, the first encoder collects the information of statistics and rate-quality characteristics. An on-line information processor then uses the collected information to derive the best coding strategy for the second encoder to encode the incoming frames in the look-ahead window. The second encoder uses the encoding parameters from the processor as the coding guide to execute the coding strategy and generate the final bitstream.

Abstract: A decoding power aware encoding method for generating a predictively encoded data stream, in which predictions, that result in a reduction in the amount of data transferred from the secondary memory to primary memory during the decoding process, are favored, said method for favoring certain predictions comprising: a model for transfer of data from secondary memory to primary memory in the decoding process; a scheme for weighting the relative merits of favoring a certain prediction and the associated loss in compression gain; and based on said weighting scheme, choosing a particular prediction from the candidates allowed by the compression scheme.

Abstract: A method and apparatus is disclosed for presenting information to a user of an audiovisual appliance such as a television receiver, a set top box, or a DVD or a VCR player. The method includes extracting source text from a graphical representation of the source text in an associated audiovisual program; translating the extracted source text into a target text in a language different than a language of the source text; and presenting the target text in a human perceptible format in a synchronized manner with the associated audiovisual program. Translating the extracted source text may include an initial step of identifying the source language.