Abstract: A method for encoding video data to reduce blocking artifacts is provided. The method initiates with identifying a macro-block as being associated with a blocking artifact. For example the macro-block may be identified as a low activity macro-block or a high activity macro-block. Then, blocking artifacts introduced through a compression operation are mitigated by adjusting both a quantization parameter and an amount of bits generated from the quantization parameter. A video encoder and a system for processing image data associated with block based compression are provided.

Abstract: A method for providing a delivery scheme for an audio augmented photograph is defined. The method initiates with combining digital audio data and digital image data to define an audio augmented digital image. Then, the audio augmented digital image is transmitted to a receiving device. After receiving the audio augmented digital image, the audio data is extracted. Next, an audio augmented printed image is generated, wherein the audio augmented printed image includes visually imperceptible embedded audio data. Then, detection of the embedded audio data is enabled when the audio augmented printed image is scanned. A computer readable media, an image delivery system and devices configured to augment digital image data with audio data and transform an audio augmented digital photograph to an audio augmented printed photograph are also provided.

Abstract: An efficient compressed-domain-based algorithm that works by combining the IDCT and pixel-domain averaging into a single compressed domain operation, where a part of the cost of even this simplified operation can be absorbed into the de-quantization process. The original image need not be decompressed fully, only to the point of extracting the transform coefficients. Compressed-domain scaling for all scales k/8 for 1≦k≦8, including ⅜, ⅝, ¾, ⅞, is provided. By chaining multiple operations, any scale factors of the form k1k2 . . . kl/8l for integers ki's in the range [1,7] can be achieved.

Abstract: A power-scalable hybrid technique to reduce blocking and ringing artifacts in low bit-rate block-based video coding is employed in connection with a modified decoder structure. Fast inverse motion compensation is applied directly in the compressed domain, so that the transform (e.g., DCT) coefficients of the current frame can be reconstructed from those of the previous frame. The spatial characteristics of each block is calculated from the DCT coefficients, and each block is classified as either low-activity or high-activity. For each low-activity block, its DC coefficient value and the DC coefficient values of the surrounding eight neighbor blocks are exploited to predict low frequency AC coefficients which reflect the original coefficients before quantization in the encoding stage. The predicted AC coefficients are inserted into the low activity blocks where blocking artifacts are most noticeable.

Abstract: A method for performing inverse memory compensation is provided. The method initiates with receiving a video bit stream. Then, a transform matrix type is identified. The transform matrix type is either a half pixel matrix and a full pixel matrix. If the transform matrix type is a half pixel matrix, then the method includes applying a factorization technique to decode the bit stream corresponding to the half pixel matrix. If the transform matrix type is a full pixel matrix, then the method includes applying an integer approximation technique to decode the bit stream corresponding to the full pixel matrix. A computer readable media, a printed circuit board and a video decoder for performing inverse motion compensation are also provided.

Abstract: A method for reducing the memory requirements for decoding a bit stream is provided. The method initiates with receiving a video bit stream. Then, a frame of the bit stream is decoded into a discrete cosine transform (DCT) domain representation. Next, non-zero coefficients of the DCT domain representation are identified. Then, a hybrid data structure is assembled. The hybrid data structure includes a fixed size array and a variable size overflow vector. Next, the non-zero coefficients of the DCT domain representation are inserted into the hybrid data structure. A computer readable media, a printed circuit board and a device configured to decode video data are also provided.

Abstract: A method and apparatus for receiving and processing digitized video data in a discrete cosine transform (DCT) domain exploits the orthogonality of a convolution function along with the data sparseness present in the DCT domain. In a preferred method, the convolution function is applied to vectors associated with a plurality of input video data frames in consideration of an orthogonal characteristic of the convolution function. More specifically, cross-products of the input vectors which would yield non-zero output vectors are identified and weighting factors for the cross-products are determined in consideration of this orthogonality. An output video data frame is generated from the non-zero output vectors and the weighting factors. A convolution operation based upon the disclosed method and aparatus is suitable for chroma-keying as well as other video or audio blending applications.

Abstract: Multiplier-free implementation of an approximation of the DCT used in image and video processing. In accordance with the primary aspect of the present invention, image and video processing is done with no multiplications and a fewer number of operations through the application of a modified Arai, Agui, and Nakajima (AAN) scheme for eight-point DCT.

Abstract: An original image is sharpened by obtaining a first frequency-domain representation of the original image, selecting one or more elements from this first representation based on one more criteria such as element magnitude and frequency, scaling the selected elements according to one or more scale factors, and forming a second frequency-domain representation by combining the scaled selected elements with the unselected elements of the first representation. A sharpened reproduction of the original image may be generated by applying an inverse transform to the second frequency-domain representation. A technique for deriving the value of the one or more scale factors is also discussed.

Abstract: Image processing techniques which involve direct manipulation of the compressed domain representation of an image to achieve the desired spatial domain processing without having to go through a complete decompression and compression process. The techniques include processing approaches for performing the eight operations in D4 (the dihedral group of symmetries of a square) on JPEG images using the discrete cosine transform (DCT) domain representation of the images directly. For a task such as image rotation by 90° (an operation in D4), DCT-domain based methods can yield nearly a five-fold increase in speed over a spatial-domain based technique. These simple compressed-domain based processing techniques are well suited to the imaging tasks that are needed in a JPEG-based digital still-camera system.

Abstract: Disclosed is an error-correction method and apparatus for transmission of block-based coding standard compliant video data, such as H.263, MPEG or JPEG. A picture is divided into slices wherein each slice of the picture is coded into a group of blocks (GOB) in which each macroblock in the GOB is encoded only with reference to other macroblocks in the same GOB. An erasure slice, which is also a GOB, is then formed wherein the data for each macroblock of the erasure slice is determined by summing a corresponding macroblock in each of the GOBs containing slices of the picture. The GOBs containing slices of the picture are then transmitted, along with the GOB containing the erasure slice, as block-based coding standard compliant data packets. The result is that if any one of the GOBs containing slices of the picture is lost, then the lost GOB can be reconstructed from the remaining GOBs and the erasure slice.

Abstract: A watermarking scheme for images which includes techniques for inserting and extracting fragile watermarks in the frequency domain and for determining whether an image so watermarked has been tampered with. Watermark insertion is accomplished by embedding the bits of a digital signature of a hash function of the image in the frequency coefficients of the image. Tamper detection is accomplished generally as follows: the fragile watermark which was embedded during the watermark insertion process is extracted from the image; the hash function of the image is computed as in the insertion process; it is verified using a public key whether the extracted watermark is a valid signature of the hash value. If so, then there is assurance that the image has not been tampered with. Otherwise, there is reason to conclude that the image has been tampered with.

Abstract: An automatic testing method and device is described that can test a video sequence coder/decoder system and either assess the quality of decoded sequences or rate the fidelity of the coding chain. The device produces synthetic test patterns that induce the appearance of known artifacts, then tracks and evaluates such artifacts. Based on this evaluation, it can rate the system's performance in a way that correlates well with human assessments. In our testing device, the quality estimation module performs this function.

Abstract: A method of compressing color source image data includes forming a quantization table with a "supra-threshold" term. This method includes a step of selecting a set of target images, where each target image includes one or more image elements such as text. These image elements are then analyzed to identify those that are more important for visual quality. These "supra-threshold" terms are then selected that gives a larger weight to the quantization table elements that correspond to important image elements and a smaller weight to the table elements that correspond to less important image elements. This process selectively weights the characteristics of each DCT basis vectors. By giving larger weights to the table elements that correspond to the "up-downness" of the image, i.e., the vertical attributes of the image elements, and the "left-rightness" of the image, i.e.

Abstract: The text and image enhancing technique according to the invention is integrated into the decoding or inverse quantization step that is necessarily required by the JPEG standard. The invention integrates the two by using two different quantization tables: a first quantization table (Q.sub.E) for use in quantizing the image data during the compression step and a second quantization table used during the decode or inverse quantization during the decompression process. The second quantization table Q.sub.D is related to the first quantization table according to a predetermined function of the energy in a reference image and the energy in a scanned image. The energy of the reference image lost during the scanning process, as represented by the energy in the scanned image, is restored during the decompression process by appropriately scaling the second quantization table according to the predetermined function.

Abstract: A method is described for filtering compressed images represented in the discrete-cosine-transform (DCT) domain. The filter includes three sparse, vertical submatrices which are sparse versions of the vertical filter components (VFCs) of a desired filter function that have been combined in such a way as to eliminate many of the non-zero elements. The filter also includes three sparse, horizontal transpose submatrices, which, like the vertical submatrices, are sparse versions of the horizontal filter components of the filter function. The sparseness of these sparse submatrices yields a significant reduction in the number of computations required to filter the image in the DCT domain. To take advantage of this discovery, the input DCT data blocks are "butterflied" to retain the relationship between the input data blocks and the filtered output data blocks as a function of these sparse submatrices.

Abstract: This disclosure provides a method of block-based motion estimation used in video compression. The compression process, derives change data for a new frame of data (with respect to a reference frame) by first dividing the frame structure into data tiles (or data blocks) of identical size. Each tile in the new frame is compared to a localized window (about the tile's expected position) in the reference frame to search for a best fit, and thereby provide motion data for the particular tile. Once the best fit is determined, motion-compensated difference data is determined, and stored with the motion data for each tile to complete the process. To achieve computation efficiency, each tile under analysis is preferably converted to single-bit value data and searching and comparisons are performed based on such transformed single-bit data. The single-bit data is computed by convolving the original image data with a low-pass filter to obtain a threshold matrix.

Abstract: Downsampling and inverse motion compensation are performed on compressed domain representations for video. By directly manipulating the compressed domain representation instead of the spatial domain representation, computational complexity is significantly reduced. For downsampling, the compressed stream is processed in the compressed (DCT) domain without explicit decompression and spatial domain downsampling so that the resulting compressed stream corresponds to a scaled down image, ensuring that the resulting compressed stream conforms to the standard syntax of 8.times.8 DCT matrices. For typical data sets, this approach of downsampling in the compressed domain results in computation savings around 80% compared with traditional spatial domain methods for downsampling from compressed data. For inverse motion compensation, motion compensated compressed video is converted into a sequence of DCT domain blocks corresponding to the spatial domain blocks in the current picture alone.

Abstract: An inverse discrete cosine transform ("IDCT") implementation specifically for the decompression of JPEG, MPEG and Px64 encoded image and video data uses a preprocessing step embedded in a Huffman decoding process to classify data blocks prior to computing the IDCT. The use of data block classification, along with the use of pruned IDCTs appropriate for the specific block class, reduces the total number of multiply and addition operations necessary to decompress an encoded data block, and thereby allows faster data decompression. Synthesis of coefficients suitable for multiplication allows efficient implementation of the novel decompression technique in typical microprocessor architectures, including RISC processor architectures.