Showing papers on "Upsampling published in 1999"

Multiresolution signal processing for meshes

Abstract: We generalize basic signal processing tools such as downsampling, upsampling, and filters to irregular connectivity triangle meshes. This is accomplished through the design of a non-uniform relaxation procedure whose weights depend on the geometry and we show its superiority over existing schemes whose weights depend only on connectivity. This is combined with known mesh simplification methods to build subdivision and pyramid algorithms. We demonstrate the power of these algorithms through a number of application examples including smoothing, enhancement, editing, and texture mapping.

Reconfigurable SIMD coprocessor architecture for sum of absolute differences and symmetric filtering (scalable MAC engine for image processing)

Abstract: The proposed architecture is integrated onto a Digital Signal Processor (DSP) as a coprocessor to assist in the computation of sum of absolute differences, symmetrical row/column Finite Impulse Response (FIR) filtering with a downsampling (or upsampling) option, row/column Discrete Cosine Transform (DCT)/Inverse Discrete Cosine Transform (IDCT), and generic algebraic functions The architecture is called IPP, which stands for image processing peripheral, and consists of 8 multiply-accumulate hardware units connected in parallel and routed and multiplexed together The architecture can be dependent upon a Direct Memory Access (DMA) controller to retrieve and write back data from/to DSP memory without intervention from the DSP core The DSP can set up the DMA transfer and IPP/DMA synchronization in advance, then go on its own processing task Alternatively, the DSP can perform the data transfers and synchronization itself by synchronizing with the IPP architecture on these transfers This architecture implements 2-D filtering, symmetrical filtering, short filters, sum of absolute differences, and mosaic decoding more efficiently than the previously disclosed architectures of the prior art

Optimal filtering and upconversion in OFDM systems

Abstract: Systems and methods for converting a baseband OFDM signal to an IF signal while minimizing lengthening of the impulse response duration experienced by the OFDM signal. A conversion technique according to the present invention provides sufficient filtering to limit the effects of spurious frequency domain components caused by transitions between successive OFDM bursts. In one embodiment, the filtering is provided by a combination of a finite impulse response (FIR) filter having non-linear phase characteristics and a cyclic convolution filter. Conversion from the frequency domain into the time domain, upsampling, and cyclic filtering may be combined into one operation.

Method for motion compensated frame rate upsampling based on piecewise affine warping

Abstract: Interpolation of a new frame between a previous frame and a current frame of a video stream by motion compensated frame rate upsampling. The interpolation method includes identifying nodes and edges of objects such as triangles present in the previous frame, constructing a superimposed triangular mesh based on the identified nodes and edges, estimating displacement such nodes in the superimposed triangular mesh from the previous frame with respect to the current frame, and rendering the new frame based on the estimated displacement of nodes. Additionally, pixels of the previous frame and the current frame may be classified according to whether a pixel's value has changed from the previous frame to the current frame. This classification may be used during rendering to reduce overall processing time. Pixel-based forward motion estimation may be used to estimate motion of pixels between the previous frame and the current frame and the estimated motion may be used in estimating node displacement.

Unsampled hd MPEG video and half-pel motion compensation

Abstract: A high definition digital video presentation system is provided with an interpolator which upsamples standard resolution MPEG pictures to higher resolution, preferably to double the number of pixels in both the vertical and horizontal dimensions. The upsampling is performed after the decoding and transformation of the incoming picture and prior to the buffering of the picture data or the performance of motion compensation by which pixel data are copied from previously received pictures. In the preferred embodiment, frame pictures are decoded, upsampled and stored in full frame picture buffers, preferably, in two buffers, one to be copied into each field of subsequent pictures in the performance of motion compensation. Interpolation may be by frame, where intervening lines of pixels are interpolated from the pixels of adjacent odd and even lines, and by field, where an intervening odd (or even) line of pixels is interpolated from the pixels of each pair of adjacent even (or odd) lines. With frame interpolation, both buffers are stored with a frame interpolated upsampled picture, while with field interpolation, one buffer is stored with a picture interpolated only from even lines and one is stored with a picture interpolated only from odd lines. The decision to interpolate by frame or by field is made macroblock-by-macroblock by the processor based on an analysis of received data to provide the method that most accurately interpolates the intervening pixels. With the upsampled pictures stored in the buffers, no further half-pel interpolation is required for half-pel motion compensation of new pixels, which is carried out by directly copying original or previously interpolated pixels. Preferably pixels at original pixel positions are, however, interpolated from only original pixel data according to MPEG techniques.

Multiplexer reconfigurable image processing peripheral having for loop control

Abstract: The proposed hardware architecture is integrated onto a Digital Signal Processor (DSP) as a coprocessor to assist in the computation of sum of absolute differences, symmetrical row/column Finite Impulse Response (FIR) filtering with a downsampling (or upsampling) option, row/column Discrete Cosine Transform (DCT)/Inverse Discrete Cosine Transform (IDCT), and generic algebraic functions. The architecture is called IPP, which stands for image processing peripheral, and consists of 8 hardware multiply-accumulate units connected in parallel and routed and multiplexed together. The architecture can be dependent upon a Direct Memory Access (DMA) controller to retrieve and write back data from/to DSP memory without intervention from the DSP core. The DSP can set up the DMA transfer and IPP/DMA synchronization in advance, then go on its own processing task. Alternatively, the DSP can perform the data transfers and synchronization itself by synchronizing with the IPP architecture on these transfers. This hardware architecture implements 2-D filtering, symmetrical filtering, short filters, sum of absolute differences, and mosaic decoding more quickly(in terms of clock cycles) and efficiently than previously disclosed architectures of the prior art which perform the same operations in software.

Implementation of a New Algorithm Using the STFT with Variable Frequency Resolution for the Time-Frequency Auditory Model

Abstract: A signal processing technique is proposed for the time-frequency analysis of unsteady sound signals considering the auditory perception model and is called VFR-STFT (short-time Fourier transform with variable frequency resolution). Conventional STFT, which is commonly used for the spectral analysis of unsteady sounds, is not suitable for the auditory model because the frequency resolution of the spectral analysis within the hearing system is not constant but varies with frequency. The frequency resolution of the VFR-STFT can be adjusted to a number of analyzed frequency ranges by introducing the downsampling technique. With the VFR-STFT, calculation schemes are presented for minimizing undesirable effects, such as the distortion of the overall sound level due to nonoverlapping of the analysis windows and the impairment of partial spectra due to the finite order of antialiasing filters. In addition a procedure for equalizing time grids at all frequency ranges is included in order to describe the two-dimensional time-frequency map (TFM) having different time grids. The proposed VFR-STFT is applied to the spectral analysis of the extraction of tonal components in an unsteady sound. The results are compared to those from other time-frequency analysis methods such as STFT, VFR-FFT (fast Fourier transform with variable frequency resolution), and the wavelet packet method.

A fast scheme for downsampling and upsampling in the DCT domain

Abstract: Given a video frame or image in terms of its 8×8 block-DCT coefficients, we wish to obtain a downsized or upsized (by factor of two) version of this frame also in terms of 8×8 block-DCT coefficients. We propose an algorithm for achieving this directly in the DCT domain which is computationally much faster, produces visually sharper images and gives significant improvements in PSNR (typically 4 dB better), compared to other compressed domain methods based on bilinear interpolation. The downsampling and upsampling schemes combined together preserve all the low-frequency DCT coefficients of the original image. This implies tremendous savings for coding the difference between the original (unsampled image) and its prediction (the upsampled image). This is desirable for many applications based on scalable encoding of video.

Upsampling filter for a down conversion system

Abstract: An HDTV down conversion system including apparatus for forming a set of low resolution down-sampled pixel values corresponding to a current frame of a video signal from a set of low resolution pixel values corresponding to a residual image of a current frame of the video signal and from a set of down-sampled low resolution pixel values corresponding to reference frames of the video signal. The apparatus includes a memory for storing the set of down-sampled low resolution pixel values. An up-sampling processor receives from the memory and uses Lagrangian interpolation to convert the set of down-sampled low resolution pixel values corresponding to the reference frame of the video signal into a set of up-sampled low resolution pixel values corresponding to the reference frame of the video signal. A summing processor adds the set of low resolution pixel values corresponding to the residual image of the current frame of the video signal to the set of up-sampled low resolution pixel values corresponding to the reference frame of the video signal to form a set of low resolution pixel values corresponding to the current frame of the video signal. A decimating processor deletes selected ones of the set of low resolution pixel values corresponding to the current frame to generate the set of low resolution down-sampled pixel values corresponding to the current frame of the video signal.

Filtering video signals containing chrominance information

Abstract: The present invention relates to filtering an interlaced input digital signal containing fields of chrominance information preparatory to converting the format of the signal by means of a downsampling conversion from a 4:2:2 format to a 4:2:0 format. In the invention, the input signal is applied to a delay circuit to derive samples of the input signal representing spatially separated elements from each chrominance field where the spatial separation is of one line. The magnitudes of the samples are compared relative to one another to identify frequencies which fall within different high and low frequency ranges. An adaptive filter has a plurality of frequency responses corresponding to the frequency ranges and a frequency response is selected in accordance with the identified frequency range of the input signal samples.

Single-sided subband filters

Abstract: In a method of processing an input signal, the input signal is divided into a plurality of subbands with the aid of bank of complex valued, single-sided subband filters. The single-sided frequency spectra of the resulting subbands make aliasing negligible at near twice the critical downsampling rates.

Non-uniform relaxation procedure for multiresolution mesh processing

Abstract: An irregular connectivity mesh representative of a surface having an arbitrary topology is processed using a non-uniform relaxation procedure. The non-uniform relaxation procedure minimizes differences between vectors normal to faces of pairs of triangles having a common edge and located within a designated neighborhood of a given vertex. The relaxation procedure may be used to construct subdivision and pyramid algorithms for performing processing operations such as upsampling, downsampling and filtering on irregular connectivity meshes. The signal processing algorithms may be utilized in applications such as smoothing, enhancement, editing, texture mapping and compression.

Time discrete filter

Abstract: A decimating time discrete filter comprises transform means ( 2 ) for transfornming a real valued input signal i into a complex signal u represented by its real part Re{U} and its imaginary part Im{u}. The output signal u is down sampled by means of a decimator ( 4 ) which reduces the sample frequency by a factor of 2. Due to the downsampling process, a desired tone input signal can be translated to a different frequency. In order to prevent interference from signals at the input having the same frequency as the desired input signal, the transform means ( 2 ) have a transfer function with an increased width transition band from the frequency of the desired input signal to the frequency to which said input frequency is converted.

A fast scheme for altering resolution in the compressed domain

Abstract: Given a video frame or image in terms of its 8/spl times/8 block-DCT coefficients we wish to obtain a downsized (lower resolution) or upsized (higher resolution) version of this frame also in terms of 8/spl times/8 block -DCT coefficients. We propose an algorithm for achieving this directly in the compressed domain which is computationally much faster, produces visually sharper images and gives significant improvements in PSNR (typically 4 dB better compared to other compressed domain methods based on bilinear interpolation). The downsampling and upsampling schemes combined together preserve all the low-frequency DCT coefficients of the original signal. This implies tremendous savings for coding the difference between the original (unsampled image) and its prediction (the upsampled image). This is desirable for many applications based on scalable encoding of video.

A spatial interpolation algorithm for the upsampling of uniform linear arrays

Abstract: A method to spatially interpolate (upsample) a uniform linear array (ULA) originally sampled at less than the spatial Nyquist rate in the presence of noise is considered. The multichannel narrowband interpolation algorithm is based on a forward-backward linear prediction approach. The method interpolates N-1 virtual sensor outputs based on the outputs of N real sensors. This is done by solving a set of forward-backward linear prediction equations that relate the known (real) sensor outputs to the unknown (virtual) sensor outputs. Simulation results that validate the performance and usefulness of this method are presented in the context of code division multiple access (CDMA) mobile communications. The algorithm is applied to increase the user capacity and resolution and decrease the mutual coupling of ULA's used in conjunction with cellular CDMA systems.

A VLSI architecture for a frequency-agile single-chip 10-MBaud digital QAM modulator

Abstract: An architecture for VLSI implementation of an all-digital frequency-agile single-chip quadrature amplitude modulation (QAM) modulator with a continuously-variable symbol rate and intermediate frequency (IF) has been investigated. The proposed chip architecture accommodates a variable IF center frequency up to 70 MHz and a programmable symbol rate from a few kHz up to 10 MHz. The single-chip modulator consists of a square-root Nyquist filter, two halfband filters to cancel the images from upsampling, a variable interpolator to accommodate continuously selectable symbol rates, and a frequency translator. This paper discusses the architecture techniques used to achieve a 200 MHz throughput rate in a modest CMOS technology.

Complex carrier signal generator for determining cyclic wave shape

Abstract: A method for generating digital carrier signals for application as the carrier to a digital modulator includes providing a first repeating sequence of complex values occurring at a given sample rate and upsampling these values to a higher sample rate. A second repeating sequence of complex values is provided, wherein respective complex values in the second repeating sequence occur at the higher sample rate. The second sequence of complex values is employed to modulate the upsampled first sequence of complex values and thereby provide the complex carrier signal.

Role of over-sampled data in superresolution processing and a progressive up-sampling scheme for optimized implementations of iterative restoration algorithms

Abstract: Super-resolution algorithms are often needed to enhance the resolution of diffraction-limited imagery acquired from certain sensors, particularly those operating in the millimeter-wave range. While several powerful iterative procedures for image superresolution are currently being developed, some practical implementation considerations become important in order to reduce the computational complexity and improve the convergence rate in deploying these algorithms in applications where real-time performance is of critical importance. Issues of particular interest are representation of the acquired imagery data on appropriate sample grids and the availability of oversampled data prior to super-resolution processing. Sampling at the Nyquist rate corresponds to an optimal spacing of detector elements or a scan rate that provides the largest dwell time (for scan- type focal plane imaging arrays), thus ensuring an increased SNR in the acquired image. However, super-resolution processing of this data could produce aliasing of the spectral components, leading not only to inaccurate estimates of the frequencies beyond the sensor cutoff frequency but also corruption of the passband itself, in turn resulting in a restored image that is poorer than the original. Obtaining sampled image data at a rate higher than the Nyquist rate can be accomplished either during data collection by modifying the acquisition hardware or as a post-acquisition signal processing step. If the ultimate goal in obtaining the oversampled image is to perform super- resolution, however, upsampling operations implemented as part of the overall signal processing software can offer several important benefits compared to acquiring oversampled data by hardware methods (such as by increasing number of detector elements in the sensor array or by microscanning). In this paper, we shall give a mathematical characterization of the process of image representation on a sample grid and establish the role of oversampling by studying the dynamics of information transfer during image restoration. A new progressive upsampling procedure is presented that provides optimized implementations of iterative superresolution. Finally, the super-resolution performance of the overall scheme that combines the progressive upsampling technique with a maximum likelihood restoration algorithm will be demonstrated quantitatively by presenting processed passive millimeter-wave imagery data.© (1999) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

A Fast Scheme for Downsampling and Upsampling in the DCT

Abstract: Given a video frame or image in terms of its 8 x 8 block-DCT coefficients we wish to obtain a downsized or upsized (by factor of two) version of this frame also in terms of 8 x 8 block-DCT coefficients. We pro- pose an algorithm for achieving this directly in the DCT domain which is computationally much faster, produces visually sharper images and gives significant improvements in PSNR (typically 4 dB better) com- pared to other compressed domain methods based on bilinear interpolation. The downsampling and upsam- pling schemes combined together preserve all the low- frequency DCT coefficients of the original image. This implies tremendous savings for coding the difference between the original (unsampled image) and its pre- diction (the upsampled image). This is desirable for many applications based on scalable encoding of video.

Method and apparatus for sampling frequency conversion

Abstract: There are three different sampling frequencies used in MPEG1 audio, namely 32kHz, 44.1kHz and 48kHz. DVD players are most often sampled at 48kHz, but audio CD players are sampled at 44.1kHz. If such devices are connected to e.g. a digital audio amplifier, and generally for MPEG processing, a corresponding sampling frequency conversion is required. The sampling frequency conversion is carried out using a chain of upsamplers and downsamplers together with proper lowpass filtering, wherein the single up- or downsampling stages do use small conversion ratios.