Abstract: In one embodiment, an edge or border processor modifies the undisplayed edges of an uncompressed video input signal before it is applied to a compression engine. This configuration reduces the number of bits used in the compression process. The undisplayed edges of the video signal are modified either by being turned black, by blurring, or by copying, depending on the application. As an aspect of this embodiment, the edge processing functions are integrated into the compression engine or video encoder. In another embodiment, edges of the uncompressed video input signal are not preprocessed prior to encoding the signal. Instead, a video encoder includes a motion estimator with special edge processing that, e.g., uses only zero motion vectors for edge macroblocks. Further, the video encoder includes a quantization processor having a special edge processor. In either embodiment, the configuration of the output bit stream is compatible with the MPEG-2 standard.

Abstract: A method and system are provided for real-time special effects processing, compressing and storing video signals. A memory is provided for storing plural frames including a sequence of one or more uncompressed frames. One or more processors are also provided for forming a sequence of special effect frames. Each special effect frame is a combination of one uncompressed frame (stored in the memory) and frame data other than the uncompressed frame (which combination can be a linear combination). The processor(s) also compresses the special effect frame.

Abstract: A method is provided for transcoding, i.e., decoding, and re-encoding, a previously encoded video signal, to a second encoded representation. A number k>1 of previously encoded pictures of the previously encoded video signal are received in a buffer. Each of the k previously encoded pictures in the buffer is scanned to gather information on each of the k previously encoded pictures. An encoding parameter is allocated to a first one of the k previously encoded pictures which precedes each other one of the k previously encoded pictures in encoded order of the previously encoded video signal. The encoding parameter is allocated based on the information gathered for each of the k previously encoded pictures. The first previously encoded picture is decoded to produce a decoded picture. The decoded picture is re-encoded to generate a re-encoded picture in a second encoded representation of the video signal. The re-encoding is performed in a fashion which depends on the encoding parameter allocated thereto.

Abstract: A method and system are provided for capturing and compressing an original uncompressed video signal which enables decoding and reversible reconstruction of a decompressed version of the original video signal. The system includes an input for receiving a signal indicating a special effect operation by which a first frame of a video signal is irreversibly transformed to a special effect frame. This is achieved by combining decompressed frame pixel data of the first frame with information comprising either pixel data of a second frame or a single scaling value to be used to scale plural pixels of a frame. The information indicates a special effect operation which can be performed on decompressed pixel data of the first frame to produce a special effect frame (e.g., the information can include a separate indicator specifying a specific operation or the information by its mere presence can indicate the special effect operation). The system also includes a processor for compressing pixel data of the first frame.

Abstract: A method and apparatus for increasing the processing speed or decreasing circuit complexity (or both) of a digital image processor by eliminating the conventional multiplication operation from the RGB to YUV and YUV to RGB transforms having NTSC compatible equation coefficients. In particular, the change of color-basis matrix for converting RGB data to obtain YUV data is factored into a scale factor and a coefficient matrix. The coefficient matrix contains coefficients such that its product with the RGB input vector can be performed using only seven binary additions and five shift operations. This process effectively removes the conventional multiplication operations from the RGB to YUV conversion. The scale factor is conveniently absorbed into the quantization of the YUV data before transmission. The quantization process includes a multiplication already and the two multiplication steps are folded into one step.