Abstract: A video coding technique having motion compensation for bi-directional predicted frames (B-frames) and predicted frames (P-frames). The video coding technique involves a single-loop prediction-based base layer which includes base layer B- and/or P-frames that are generated from “extended” or “enhanced” base layer reference frames during base layer coding.

Abstract: A method and device is disclosed for encoding a group of video frames. A first number of frames in the group is motion compensated temporally filtered based on multiple frames. A second number of frames in the group is motion compensated temporally filtered based on single reference frames. The first and second number of frames are transformed into wavelet coefficients. Further, the wavelet coefficients are entropy encoded. In one example, the first number of frames are located within a first portion of the group and the second number of frames is located within a second portion of the group. In another example, the first number of frames and the second number of frames are interleaved within the group.

Abstract: An improved multiple description coding (MDC) method and apparatus is provided which extends multi-description motion compensation (MDMC) by allowing for multi-frame prediction and is not limited to only I and P frames. Further, the coding method of the invention extends MDMC for use with any conventional predictive codec, such as, for example, MPEG2/4 and H.26L. The improved MDC permits the use of any conventional predictive coder for use as a top and bottom predictive encoder. Further, the top and bottom predictive coders can advantageously include B-frames and multiple prediction motion compensation. Still further, any of the top, middle and bottom predictive encoders can be a scalable encoder (e.g., FGS-like or data-partitioning like where the motion vectors (MVs) are sent first, temporal scalability etc.).

Abstract: The present invention is directed to a method and device for encoding a group of video frames. According to the present invention, regions in at least one frame in the group are matched to regions in multiple reference frames. The difference between pixel values of the regions in the at least one frame and the regions in the multiple reference frames is calculated. The difference is transformed into wavelet coefficients. The present invention is also directed to a method and device for decoding a group of frames by performing the inverse of the above described encoding.

Abstract: A circular buffer, i.e., a chain of buffers forming a circle, is provided for managing packet loss detection in Internet streaming. The detection latency is determined by the size of the buffer chain, which can be dynamically adapted to network conditions and application requirements. The present invention can achieve reasonable detection accuracy.

Abstract: A method for streaming enhancement layer video frame data on a variable bandwidth network involves coding original uncoded video data with a non-scalable codec to generate I and P base layer frames; generating residual temporal B frames and SNR I and P frames from the original uncoded video data and the base layer frames; coding the temporal and SNR frames with a scalable codec; determining an available bandwidth of the variable bandwidth network; transmitting at least portions of the scalable coded temporal frames when the available bandwidth is a first bandwidth ranging between a minimum bit-rate allocated for transmitting only base layer frames and a predetermined bit-rate which is below a maximum bit-rate of the network; and additionally transmitting remaining portions of the scalable coded temporal frames and at least portions of the scalable coded SNR frames when the available bandwidth is a second bandwidth extending between the predetermined bit-rate and the maximum bit-rate of the network.

Abstract: The present insertion utilizes a data relationship between B-frame motion vectors (k(B)) and P-frame motion vectors (k(P)) to simplify merging and dividing of multiple descriptions (22, 24) at network nodes (28) such as gateways by avoiding the need to decompress and re-compress at least one of the multiple descriptions.

Abstract: A method and device for coding video where a video signal is spatially decomposed into at least two signals of different frequency sub-bands, an individualized motion compensated temporal filtering scheme is applied to each sub-band signal adaptively according to signal contents, and texture coding is applied to each of the motion compensated temporally filtered subband signals adaptively according to the signal content.

Abstract: A system and method provides a realistic architecture and specifies the protocols that are necessary for carrying out adaptive and efficient protection, thereby allowing applications to dynamically switch between different protection strategies. Protection is uniquely achieved by providing the protection track (206, 208, 210) as a separate stream from the media data stream (202, 204). In this manner, changing the protection data can change the protection level or strategy, but the protection procedures remain the same. Further, the method uses Real-time Transport protocol (RTSP) instead of Internet Group Management protocol (IGMP) which can achieve faster protection and provide more flexibility to applications.

Abstract: Video coding method and device for reducing the number of motion vector bits, the method and device differentially coding the motion vectors at each temporal decomposition level by predicting the motion vectors temporally and coding the differences.

Abstract: A system and method is provided to facilitate the transmission of scalable coded video over IP networks. A proposed pre-processing method is disclosed, referred to as multi-track hinting, efficiently structures layered video into a flexible format so that it can be easily streamed over packet-switching networks in accordance with changing network conditions, complexity constraints and user preferences. A general purpose MPEG server, without major modification, is capable of automatically using multiple channels (i.e., RTP connections), thereby providing the streaming system the flexibility to adapt to changing network conditions, complexity constraints and user preferences by adjusting the number of scalable layers to be transmitted. Accordingly, the multi-track hinting method extends the functions of standard Internet streaming protocols (RTSP, SDP) to enable flexible adaptation.

Abstract: In a layered encoding system having at least one layer comprising a plurality of sub-layers (272, 274, 276), a method is disclosed herein for encoding a video image (200) composed of a plurality of pixel blocks containing at least one area determined to be significant (200, 215, 220) within a corresponding sub-layer (272, 274, 276). The method comprises the steps of; associating a level of significance with each block (250, 252) of a known size within the at least one significant area (200), associating a level of significance with successively larger blocks (222, 244) dependent upon the level of significance of at least one of the blocks (250, 252) of a known size contained within said larger block (222, 244), and mapping each of the associated levels of significance. In another embodiment of the invention, the significance map is transmitted and corresponding image layers may be reconstructed using the significance map.

Abstract: A method and apparatus for coding video includes constructing motion-compensated extended reference frames from at least portions of temporal residual image frames. The motion-compensated extended reference frames are used for bi-directionally and/or uni-directionally predicting additional levels of FGS motion-compensated residual image or temporal frames.

Abstract: A method and apparatus for coding video includes constructing motion-compensated extended base layer reference frames from base layer frames and at least portions of base layer residual image frames. The motion-compensated extended base layer reference frames are used for bi-directionally or uni-directionally predicting FGS motion-compensated residual image or temporal frames.

Abstract: A video coding technique having motion compensation within a fine granular scalable coded enhancement layer. In one embodiment, the video coding technique involves a two-loop prediction-based enhancement layer including non-motion-predicted enhancement layer I- and P-frames and motion-predicted enhancement layer B-frames. The motion-predicted enhancement layer B-frames are computed using: 1) motion-prediction from two temporally adjacent differential I- and P- or P- and P-frame residuals, and 2) the differential B-frame residuals obtained by subtracting the decoded base layer B-frame residuals from the original base layer B-frame residuals. In a second embodiment, the enhancement layer further includes motion-predicted enhancement layer P-frames.

Abstract: The present invention is directed to rearranging the transmission order of the enhancement-layer frames. By making the display and transmission order of the enhancement layer frames identical, a frame memory is not required on the decoder-side to hold the enhancement-layer frame until being displayed since the display can take place immediately after the decoding. Reducing the amount of memory is desirable for mobile applications or other low-power consumption devices.

Abstract: The invention relates to a method and system for modulating an MPEG-4 FGS compressed video stream for variable-bandwidth transmission. A source of a video stream is provided, and a display having means for generating a control input signal for controlling one or more of video compression, error correction code generation, or symbol constellation mapping. The source video stream is compressed using MPEG-4 FGS compression to generate a compressed output video stream. The video compression is optionally performed based upon the control input signal from the display. Error correction code for the compressed video stream may be added, and its generation and addition may be optionally controlled based upon the control input signal from the display. The resulting video symbols may then be mapped to constellations, with this mapping also optionally controlled based upon the control input signal. The resulting compressed output video stream is then transmitted to the display.

Abstract: A method and apparatus for minimizing prediction drift at low bitrates in a fine granular scalable video coding scheme that utilizes motion compensation in an enhancement layer. The method and apparatus measures motion activity within at least a portion a video; determines whether the measured motion activity is below a predetermined threshold value; codes the portion of the video with the fine granular scalable video coding scheme that utilizes motion compensation in the enhancement layer if the measured motion activity is below the predetermined threshold value; and codes the portion of the video with a fine granular scalable video coding that does not utilize motion compensation in the enhancement layer if the measured motion activity is above the predetermined threshold value.

Abstract: A system and method for generating a frequency weighted (FW) matrix for use in a Fine-Granularity-Scalability (FGS) video coding system. The system comprises: a system for plotting the average discrete cosine transform (DCT) residuals versus the zigzag DCT scan line locations for a sample video frame encoded both at a predetermined base layer bit-rate and at approximately three times the predetermined base layer bit-rate; a system for generating the difference plot of DCT residuals versus the zigzag DCT scan line locations for the video frame encoded at both the predetermined base layer bit-rate and at approximately three times the predetermined base layer bit-rate; and a system for matching and normalizing a staircase curve to the average difference plot, wherein the staircase curve values can be further mapped into the weights for the FW matrix.

Abstract: A method and system for high resolution formatting of video images and dynamically adjusting the transmission resolution of the high-resolution images is presented. The method first downscales the high-resolution images and encodes the downscaled images into base layer frames. Quality enhancement layer data is generated from the downscaled video images and the encoded data contained in corresponding base layer frames. The quality enhancement layer data is encoded into quality enhancement layer frames. The data contained in the base layer frames and corresponding quality layer frames are then upscaled and spatial scalability data is determined from the upscaled data and the original image. The spatial scalability data is then encoded into spatial scalability data. During transmission of the encoded video image, each available encoded frame is transmitted using different amounts or portions of the enhancement layers so as to occupy the available bandwidth.