Abstract: Herein described is a system and method of detecting 2:2 and/or 3:2 pull-down video. In addition, aspects of the present invention provide a system and method to identify and lock onto a field phase of a received 2:2 or 3:2 pull-down video signal. The system for detecting a field phase in 2:2 pull-down video comprises a first circuitry that computes unexpected motion values, a second circuitry that implements an alternating straight through or crossover function, a third circuitry that performs processing and generates one or more control signals, a pair of counters, and a phase selection circuitry. The method comprises computing luminance approximations of pixels associated with an output video frame, computing one or more first differences, and computing a second difference for each absent pixel of the output video frame based on one of five field phases.

Abstract: Herein described are at least a system and a method of more effectively detecting and correcting irregularities when performing an inverse telecine deinterlacing of pull-down video. These irregularities may be caused by editing of the video, which when left undetected, may result in displaying video of unacceptable quality. In a representative embodiment, the method determines a first pixel luminance of an absent pixel using a luminance generated from one or more first present pixels, determines a second pixel luminance of the absent pixel using a luminance generated from one or more second present pixels, and determines a first luminance approximation for the absent pixel. In a representative embodiment, the system comprises a memory, a software resident in the memory, and a processor used to execute the software, such that executing the software performs the inverse telecine deinterlacing to result in an improve displayed video image quality.

Abstract: Various aspects of the low cost line buffer system allow a reduction in circuitry versus conventional approaches to line buffer design. A plurality of line buffers such that the output of one line buffer in the plurality of line buffers may be coupled to an input of a succeeding line buffer in the plurality of line buffers. A first line buffer in the plurality of line buffers may be coupled to an input write data signal, while the width of a subsequent plurality of line buffers may be less than or equal to the width of the previous line buffers in the plurality of line buffers.

Abstract: In a video system, a method and system for control of a multi-field deinterlacer including visually pleasing start-up and shut-down are provided. During a hard start, a deinterlacer may deinterlace a first received field and provide the resulting frame for display until sufficient fields are received to generate a pixel constellation for deinterlacing subsequent received fields. The deinterlacer may also provide a constant frame for display until sufficient fields are received to generate the pixel constellation. During normal operation, the deinterlacer may mitigate the effects of a discrepancy as to the received field type. When a video source is jittery or there is uncertainty as to the field type, deinterlacing may be forced to be based on spatial information. When a video source is shut off, a portion of the remaining stored fields may be flushed. At least one of the flushed fields may be deinterlaced and displayed.

Abstract: In a video system, a method and system for quantized historical motion for motion detection in a motion adaptive deinterlacer are provided. A motion-adapted value for an output pixel in a current output field may be determined based on a current motion value and a historical motion value. The current motion value may be based on a pixel constellation and may be quantized once determined. The quantized current motion value may be stored and may be used to determine a quantized historical motion value for an output pixel in a field occurring after the current field. At least one quantized historical current motion value may be used to determine the quantized historical motion value for the output pixel in the current output field. The quantized historical motion value may be reverse quantized to determine the historical motion value for the output pixel.

Abstract: In a video system, a method and system for pixel constellation in a motion detection deinterlacer are provided. An output pixel in an absent line of a current frame may be determined based on a constellation of pixels from at least one present line above and at least one present line below the absent line. The present lines may be immediately above and/or below the absent line. Pixels from the present lines immediately above and/or below include the pixel directly above and/or below the output pixel and/or at least two pixels to either side. The output pixel may also be determined from present lines in fields that occur prior to and/or after the current field. The fields may be those that occur immediately before and/or after the current field and/or the second and third field that occur after the current field.

Abstract: Aspects of the present invention may be found in a system and method to more effectively detect and correct irregularities of video in a pull-down video signal, such as a 3:2 (or 2:2) pull-down video signal. Various aspects of the present invention provide a system and method to determine when the cadence of pull-down video has changed. This may occur, for example, when a bad-edit is encountered in the pull-down video. Other aspects of the present invention may be found in a system and method of detecting and correcting distortion or artifacts occurring at the pixel level. The distortion or artifacts are detected and corrected at a per-pixel level while performing a reverse pull-down of the video signal. Finally, aspects of the invention allow the automatic enabling of per-pixel correction when performing a reverse 3:2 directed weave. A computed variable is used to automatically activate the per-pixel correction.

Abstract: In a video system, a method and system for minimizing both on-chip memory size and peak DRAM bandwidth requirements for multi-field deinterlacers are provided. Present line pixels and absent line pixels may be determined and may be ordered into upper line pixels and lower line pixels based on whether an interlaced output frame is top field or bottom field originated. Upper line pixels may be buffered in an upper line FIFO that may hold half a video line and lower line pixels may be buffered in an lower line FIFO that may hold a video line. A source switch and a line length counter may be used to assemble the deinterlaced output frame by selecting from the buffered upper line pixels and the buffered lower line pixels. The deinterlaced output frame may be buffered in an output FIFO.

Abstract: Certain aspects of the invention may comprise a plurality of algorithms and architectures capable of performing 3:2 pull-down cadence detection and reverse 3:2 pull-down of a received video signal. In one embodiment, a method of identifying a pull-down field in a pull-down video stream comprises generating one or more temporally sequential variances, correlating the one or more temporally sequential variances to one or more temporal variance patterns associated with the pull-down video, wherein each of the one or more temporal variance patterns is associated with one or more unique field phases. In one embodiment, a system for identifying a pull-down field in a pull-down video stream comprises a memory, a first circuitry capable of computing a first equation, a second circuitry capable of computing a second equation, a third correlation circuitry, one or more counters, and a fourth circuitry capable of selecting a field phase.

Abstract: A system and method that scales interlaced video fields with different sizes in a deinterlacer. The deinterlacer may expect video fields of a certain size. The input to the system may be a video stream of interlaced content, with fields having different sizes depending on the video content. The method scales the input video fields to the appropriate size expected by the deinterlacer. The expected size of the fields may be programmable. The method may provide an output with the expected size of black pixels in the absence of an input to the system.

Abstract: A system that may be used to generate documents and for broader uses such as assembling computer-processable components into computer-processable end products. In one form, the system is a document generation system having an assembly facility configured to be coupled to an origination platform; a knowledge base configured to be coupled to the assembly facility and to store objects in an object-relational hierarchy; and a content management system configured to be coupled to the knowledge base, where the content management system is configured to include an object and a rule. The system may be configured to interact with a knowledge base to create a first set of end products, each end product containing an object; apply precedence to the first set of end products; extract rules from the knowledge base; and assemble a second set of documents based upon applying precedence and rules.

Abstract: A knowledge base configured to store data structure components as objects in an object-relational hierarchy, each object configurable to have precedence, to include one or more rules, and to include content. The knowledge base may be coupled to a data structure assembly facility capable of retrieving one or more data structure components from the knowledge base and generate a tree having a root node, processing the tree beginning at the root node, and overriding objects of low precedence with objects of higher precedence.

Abstract: A method of generating documents. In one embodiment the method involves establishing an architecture for a set of rules to be used in documents that consist of a plurality of components and creating a dynamic document structure that can resolve to one or more instances of a document and that is configured to include one or more rules based on the architecture for a set of rules. The architecture for a set of rules may be established by creating a schema having a conditions element, a choose element, an iterators element, a functions elements, and an external interface element that is configured to be resolved into a value. The method may also be configured to create a static document structure that can be resolved to one or more instances of a document that includes at least some content that is determined before and some content that is unchanged during and after a resolution process.

Abstract: A data structure generation system. The system includes a knowledge base configured to store data structure components as objects in an object-relational hierarchy. Each object is configured to have a precedence, to include one or more rules, and to include content. The knowledge base may be coupled to a data structure assembly facility configured to retrieve one or more cross-referenced data structure components from a database. The one or more data structure components are configured to have a precedence level, and may be processed in a processor to generate a tree having a root node. The tree may be processed beginning at the root node, and objects of low precedence may be overidden with objects of high precedence.