Abstract: In one embodiment according to the present invention, relative z-ordering of segments in a digital image is determined. A method comprises forward and backward motion matching of image regions to determine overlap, followed by the creation of relationships (e.g., pairwise relationships) between regions and comparing the result with the original image to determine the relative z-ordering.

Abstract: A video compression method and apparatus uses an active decoder. The corresponding encoder can produce an encoded bitstream with a greatly reduced overhead by encoding a reference frame based on the structural information inherent to the image (e.g., image segmentation, geometry, color, and/or brightness), and then predicting other frames relative to the structural information. Typically, the description of a predicted frame would include kinetic information (e.g., segment motion data and/or associated residues representing information in previously occluded areas and/or inexact matches and appearance of new information, and portion of the segment evolution that is not captured by motion per se, etc.). Because the decoder is capable of independently determining the structural information (and relationships thereamong) underlying the predicted frame, such information need not be explicitly transmitted to the decoder.

Abstract: A method and apparatus for predicting and coding motion vectors in a video compression scheme is disclosed. An ordered list of segments for a reference frame is used to create a hierarchy of segments with a plurality of levels. Motion vectors for segments in the top level are entropy coded, and these vectors are used to predict vectors for segments at the next level. Residual vectors are entropy coded to correct these predictions, and the process of prediction from above and coding residuals continues recursively down through the hierarchy of segment levels. Information about the previous motion of segments may be exploited in the prediction process. A complementary method and apparatus for decoding motion vectors is also disclosed. The same segment hierarchy is used to predict motion vectors by the same method used during encoding, and these predicted vectors are added to residual motion vectors to reconstruct the actual motion vectors.

Abstract: A process and apparatus for allocating bits between the stages of a multi-stage digital image compression scheme with quantizer feedback is disclosed. The process selects a quantization schedule (from a finite number of possibilities) for all stages that minimizes an estimate of total distortion while keeping total bit demand below a constraint; uses said schedule to actually quantize one stage of the compression scheme; selects a new quantization schedule for the remaining stages using estimates for distortion and bit demand that are updated to reflect the quantization of the previous stage; actually quantizes the next stage according to the new quantization schedule; and continues recursively until it has chosen an actual quantization scale for every stage of the compression scheme.

Abstract: A process and apparatus for identifying abrupt cuts or scene changes in any ordered sequence of images. In one specific embodiment, two or more consecutive images from a sequence are introduced to a segmenter as digital frames. The segmenter independently divides each of these frames into pixel regions or segments according to some common characteristic so that every pixel belongs to exactly one segment. A segment analysis unit then performs some statistical analysis on the segment data for each of the frames and generates composite statistics for each frame. A frame comparison unit then examines these composite statistics to determine whether these frames belong to a consistent scene of images. If the composite statistics for these frames differ sufficiently, the comparison unit declares the latter frame in the sequence to belong to a new scene. This information may then be transmitted back to the data source for the purpose of marking the scene change or for any other purpose.

Abstract: A video compression method and apparatus uses an active decoder. The corresponding encoder can produce an encoded bitstream with a greatly reduced overhead by encoding a reference frame based on the structural information inherent to the image (e.g., image segmentation, geometry, color, and/or brightness), and then predicting other frames relative to the structural information. Typically, the description of a predicted frame would include kinetic information (e.g., segment motion data and/or associated residues representing information in previously occluded areas and/or inexact matches and appearance of new information, and portion of the segment evolution that is not captured by motion per se, etc.). Because the decoder is capable of independently determining the structural information (and relationships thereamong) underlying the predicted frame, such information need not be explicitly transmitted to the decoder.

Abstract: An encoding system encodes and decodes video data using an exposed area fill processor. A nonkey frame is encoded as a set of kinetic information relating segments of a reference frame to the nonkey frame and the nonkey frame is decoded from segments of at least one reference frame, the kinetic information and results of an exposed area fill process that estimates contents of the nonkey frame not already covered by the segments and the kinetic information. The exposed area fill results can be determined at the encoder and conveyed to the decoder or the decoder can determine the results independently. The encoded video data can include an indication of which of a plurality of fill schemes are used, or the decoder can determine the indication by itself. The exposed area fill information might include bounds of areas to be filled and/or pixel values of pixels therein.

Abstract: An encoder segments frames of video data and associates metadata with segments. The metadata elements can be associated with the segments that include areas of the frame associated with the metadata elements. A motion matcher can match segments of a reference frame to pixels of a current nonkey frame being encoded when a metadata associator associates elements of metadata with segments of the segmentation of the reference frame and associates a matched segment's metadata elements with matched areas of the current frame. The metadata elements might represent actions to be taken when a user of the video data indicates a selection of an area of an image that has that metadata element associated therewith. The metadata associations can be included in the encoded video data or deduced by a decoder. The metadata associations can be independent of segment indices or other segment changes.

Abstract: Video data is encoded to form compressed video data wherein at least some of the frames are encoded as nonkey frames with reference to content of reference frames. In encoding a nonkey frame, segments of a reference frame can be matched to the pixels of the nonkey frame and related by a calculated motion vector for each segment, wherein a motion vector represents a difference in position of the image area represented by the segment. Motion vectors are predicted and the encoding of the nonkey frame can omit coding for calculated motion vectors that are sufficiently close to their predicted values and encoding differences between calculated and predicted motion vectors. Differences in values can be considered residue and coded as part of residue data. A decoder making similar predictions can decode the compressed video data even when the compressed video data contains no information about the predictions for motion vectors.

Abstract: Video data is compressed using segmentation of reference frames where the segmentation results in a hierarchy of segments. The hierarchy can be generated bottom-up, where a first set of segments is generated and then a second set of segments generated by grouping segments of the first set, based on pixel value and/or segment boundaries. The hierarchy can be generated top-down, where a first set is generated and then a second set is generated by dividing up segments of the first set, based on pixel value and/or segment boundaries. The hierarchy can be included in the compressed video data or omitted such that a decoder would have to independently generate it. The hierarchy can be two or more levels. The hierarchy can be used for editing, formatting and/or compressing frames, as well as associating metadata with elements of the frame and coding motion, residue or other kinetic information.

Abstract: An encoding system includes an encoder that encodes uncompressed video data to form compressed video data and a decoder that decodes the compressed video data to form at least an approximation of the uncompressed video data. The video data comprises a sequence of a plurality of image frames comprising key frames and nonkey frames. In encoding a nonkey frame, reference is made to segments of a reference frame. Kinetic information, such as motion vectors, relates portions of the nonkey frame to the segments of it reference frame. The information content of the nonkey frame that is not conveyed by the segmentation, the reference frame content and the kinetic information can be encoded as residue data. Each segment's residue data, or the nonkey frame's residue data, can be encoded as coefficients of basis functions known to both the encoder and the decoder.

Abstract: Motion vectors encode for differences between segments of a reference frame and pixels of a current frame being encoded when the current frame is a nonkey frame. The motion vectors are encoded by grouping one or more segments of the frame into a group of segments, generating a group motion vector for the group of segments and one or more local motion vectors, wherein a local motion vector corresponds to a segment in the one or more segments and specifies a change for the segment relative to the motion vector for the group of segments and encoding the motion vector and one or more local motion vectors. The group motion vector can be determined based on a relationship between the motion vectors for the one or more segments. The grouping can be hierarchical. The motion vectors can represent translation, z-ordering, deformation, and/or lighting data.

Abstract: An efficient method of matching a segment in one image with a segment in another image. Fourier transforms are implemented to aid in the process. In one embodiment, a method involves identifying a displacement of a segment present in a first image and a second image, the displacement representing a relative change in position of the segment between the first and second images.

Abstract: A method and apparatus for client-side detection of network congestion in a best-effort packet network comprising streaming media traffic is disclosed. Said method and apparatus provide for quality streaming media services in a congested network with constrained bandwidth over the last-mile link. A client media buffer detects at least one level of congestion and signals a server to enact at least one error mechanism. Preferred error mechanisms include packet retransmissions, stream prioritization, stream acceleration, changes in media compression rate, and changes in media resolution. Said method and apparatus allow distributed management of network congestion for networks comprising multiple clients and carrying significant streaming media traffic.

Abstract: A decoder decodes compressed video data wherein nonkey frames are decoded with reference to other frames from the video data that are reference frames. The decoder generates at least a part of a segmentation of the reference frames for use in decoding nonkey frames. A nonkey frame is regenerated using kinetic information about the current frame and the reference frame segmentation. Kinetic information might include segment translation information. Where the segmentation used in encoding the compressed video data can vary among a plurality of segmentation schemes, the decoder determines which segmentation scheme is used from selection indications in the compressed video data or from previously decoded video data. The decoder might also use partial segmentation information, segmentation hints, partial segment canonical information and/or canonical hints in its segmentation process. The decoder might also process segment-related metadata extracted the compressed video data.

Abstract: Video data is compressed with nonkey frames encoded with reference to segmentation of reference frames, where the encoded video data includes kinetic information relating segments of a reference frame to pixels of a nonkey frame and the kinetic information includes translations of segments and at least one of a z-order, a deformation and a lighting change. The segmentation performed during encoding can be included in whole or part in the compressed video data. If used, z-ordering could be relative or absolute, based on changes of occlusion of segments by other segments between the frames, based on content of other frames or based on z-order indications in the video data being compressed. Other kinetic information might include segment changes between frames such as rotation, dilation, affine transformations, nonlinear transformations defined by a set of deformation parameters, linear lighting offsets in one, two or three color planes, and/or residue information.

Abstract: An encoder includes a segmenter that segments a reference frame, assigning some or all of the pixels of the reference frame to segments based on pixel color values and/or pixel location. A kinetic information generator generates kinetic information that relates regions of a nonkey frame to segments of the reference frame as indicated by the segmentation. A decoder includes its own segmenter that segments at least a part of the reference frame extracted from the compressed video data, thereby eliminating the need for the encoder to include all of the segmentation in the compressed video data. The decoder includes a nonkey frame generator that regenerates a nonkey frame using kinetic information about the nonkey frame from the compressed video data and using at least a part of the segmentation of the reference frame generated by the decoder's segmenter. The encoder segmentation and decoder segmentation can be the same or different.

Abstract: The utilization of an non-QOS guaranteed network is envisioned within a communication network to increase bandwidth when necessary. In this system two locations are connected by two separate communications networks one QOS guaranteed QOS guaranteed network while the other non-QOS guaranteed packet based network without QOS guarantee. A smart buffering system integrates the two networks.

Abstract: The utilization of an overflow link is envisioned within a communication network to increase bandwidth. In order to increase bandwidth when necessary, an alternate overflow network is utilized to transfer data between a local data center and an individual receiving location. In this process, a portion of the data packets are transmitted to a different receiving location with available bandwidth. The packets are then transmitted from this location to the original requesting location via a shared communications link between the two receiving locations, thus increasing the bandwidth between the local center and the individual receiving location until the additional demand is met.