Abstract: A reference picture is processed to generate, first, second and third pictures. The first picture has a lower resolution than the second picture and the second picture has a lower resolution than the third picture. Using independent starting positions, first and second stage field motion estimation (motion vector) searches are preformed using the first and second pictures, respectively. By performing the first stage field motion estimation search, a first field motion vector candidate is obtained. By performing the second stage field motion estimation search, a second field motion vector candidate is obtained. Using at least one of the first and second field motion vector candidates as a starting position and the third picture, a third stage field motion estimation search is performed. Alternatively, a first stage field motion estimation search is performed using the first picture to obtain a first motion vector from a top reference field and a second motion vector from a bottom reference field.

Abstract: A statistical multiplexing apparatus and method for generating and combining a plurality of encoded video bit streams. A storage device contains pre-stored a priori statistics indicative of the encoding complexity of the video signals from which the encoded video bit streams will be generated. The pre-stored a priori statistics may include inter-pixel differences in the same picture or between multiple pictures or pre-encoding a priori statistics generated during a preliminary encoding of the video signals. Examples of pre-encoding a priori statistics include the number of bits per picture at a given quantization level, an average quantization level, picture types, scene change locations and repeat field for one or more of the video bit streams. The video signals are applied to encoders which compress the signals in accordance with bit allocation decisions generated by a statistics computer.

Abstract: A process and apparatus for encoding are provided, wherein fields of a digital signal are processed to detect repeat fields. Adjacent pairs of the non-repeated fields are organized into frames. A determination is made whether to encode each of the frames as an intraframe, a predicted frame or a bidirectionally predicted frame. The frames are encoded in a specific, predefined order relative to the order of capture of the frames and the type of frame. After each bidirectionally predicted frame that immediately precedes one of the detected repeat fields, encoding of a frame is delayed for one field time. Encoding is paused after encoding each reference frame that is the very next reference frame to be encoded after a second reference frame, which second reference frame immediately precedes one of the detected repeat fields. A process and apparatus for statistically multiplexing multiple encoded digital video signals are also provided.

Abstract: A statistical multiplexing apparatus and method for generating and combining a plurality of encoded video bit streams. A storage device contains pre-stored a priori statistics indicative of the encoding complexity of the video signals from which the encoded video bit streams will be generated. The pre-stored a priori statistics may include inter-pixel differences in the same picture or between multiple pictures or pre-encoding a priori statistics generated during a preliminary encoding of the video signals. Examples of pre-encoding a priori statistics include the number of bits per picture at a given quantization level, an average quantization level, picture types, scene change locations and repeat field for one or more of the video bit streams. The video signals are applied to encoders which compress the signals in accordance with bit allocation decisions generated by a statistics computer.

Abstract: An apparatus and method for compressing multiple resolution versions of a video signal are disclosed. A first resolution version of a video signal is applied to an input of a first video compressor and to an input of a video scaler. The first video compressor encodes the first resolution version of the video signal to generate a first compressed video bit stream. The video scaler generates a reduced resolution version of the video signal from the first resolution version. The reduced resolution version is supplied to a second video compressor and to the first video compressor. The first video compressor utilizes the reduced resolution version of the video signal in performing a hierarchical motion estimation (ME) search as part of the encoding process for the first resolution version. The second video compressor encodes the reduced resolution version to generate a second compressed bit stream.

Abstract: In accordance with an illustrative embodiment of the invention, a composite video signal is decoded as follows. A filtered and optionally re-sampled luminance component is obtained by vertical low-pass filtering the composite video signal using a vertical low-pass filter. The vertical low-pass filter may also be a re-sampler so that the output is optionally a filtered and re-sampled luminance signal. (Note that the use of a single vertical low-pass filter and re-sampler is more efficient than the prior art which uses a luma comb filter followed by a vertical filter/re-sampler to obtain the filtered and optionally re-sampled luminance component). Filtered and re-sampled chrominance components are obtained by first vertically high-pass filtering and re-sampling the composite video signal using a vertical high pass filter/re-sampler. This results in filtered and re-sampled but still quadrature modulated chroma signals.

Abstract: Methods and apparatus for performing dual prime motion estimation on video fields or frames of a video signal. A 16.times.16 motion estimator generates a same-parity match vector M.sub.-- SP and an opposite-parity match vector M.sub.-- OP for a current macroblock of a current field using a reference macroblock from each of a previous same-parity and opposite-parity field. A dual prime motion estimator receives the M.sub.-- SP and M.sub.-- OP match vectors for the current macroblock, and generates a base motion vector MV and a delta motion vector DMV for the current macroblock. In a first embodiment, the dual prime motion estimator generates MV and DMV by setting MV equal to M.sub.-- SP and then selecting DMV such that the opposite-parity vector OPV to be computed by a decoder is as close as possible to M.sub.-- OP. In a second embodiment, OPV is set equal to M.sub.-- OP and MV and DMV are selected such that MV is as close as possible to the same-parity match vector M.sub.-- SP.

Abstract: A system and method for receiving a video sequence with a given aspect ratio and displaying that sequence on a device with a different aspect ratio in letterbox form by performing letterbox type conversion immediately before display. In letterbox form, the viewer sees the entire active region of the picture but the whole screen is not used. In one embodiment a video sequence with a given aspect ratio can be displayed on a device with a different aspect ratio in letterbox or pan-scan form, at the users discretion, by performing, respectively, letterbox or pan-scan conversion immediately before display. Thus, the user can decide to use the entire screen or to see the entire picture.

Abstract: A system and method for repeat field detection for use in rate conversion and video encoding of the type contemplated by the MPEG standards. Before encoding a frame, first field in the current frame is compared to the previously occurring field of the same parity. Next, a single number (the inter-field parameter) is generated, which is a measure of the difference between the two fields. If the inter-field parameter is below a threshold a signal is generated to indicate that the current field is a repeat of the previous field of the same parity. Otherwise, the signal is generated to indicate that the field is not a repeat of the previous field of the same parity (i.e. the field likely contains at least one small area of motion).

Abstract: This invention minimizes the number of non-trivial multiplications in the DCT process by rearranging the DCT process such that non-trivial multiplications are combined in a single process step. In particular, the DCT equations for the row-column application of the DCT process on k=pq points wherein p and q are relatively prime, are factored into a permutation matrix, a tensor product between matrices having p.times.p and q.times.q points, and a matrix whose product with an arbitrary vector having pq points requires pq-p-q+1 additions and/or subtractions. The tensor product is then further factored to remove non-trivial multiplications by developing a a first factor having (pq-p-q+1)/2 non-trivial multiplications and a diagonal matrix. The diagonal matrix is not unique for any set of data. Its j,j-th elements are chosen from a subproduct of the factorization of the tensor product. Once the diagonal matrix elements are chosen the remaining first factor is developed.

Abstract: The objects of this invention are accomplished by rearranging the DCT process such that non-trivial multiplications are combined in a single process step. In particular, the DCT equations for the application of the two-dimensional DCT process on k.times.k points are factored into 1) a permutation matrix, 2) a diagonal matrix, and 3) a matrix whose product with an arbitrary vector having k.sup.2 points requires considerably fewer non-trivial multiplications. Furthermore, in some cases none of these non-trivial multiplications are nested; that is, no output of a non-trivial multiplication is ever involved in another multiplication operation. The diagonal matrix is not unique for any set of data. Once the diagonal matrix elements are chosen the remaining factors are developed. When the factorization is complete, the diagonal matrix is absorbed into the quantization step which follows the DCT process. The quantization step is the multiplication of a diagonal matrix by the DCT output data.