Signal Processing

In our previous work, we introduced Spatially Varying Transforms (SVT) for video coding, where the location of the transform block within the macroblock is not fixed but varying. In this paper, we extend this concept and present a novel method, called Variable Block-size Spatially Varying Transforms (VBSVT). VBSVT utilizes Variable Block-size Transforms (VBT) in the SVT framework, and is shown to be more preferable for coding prediction error with different characteristics than fixed block-size SVT and also the standard methods that use fixed or adaptive block sizes at fixed spatial locations. In addition, VBSVT has similar decoding complexity with fixed block-size SVT and lower decoding complexity compared to standard methods as only a portion of the prediction error needs to be decoded. Experimental results show that, VBSVT achieves 4.1% gain over H.264/AVC on average over a wide range of test set. Gains become more significant at high quality levels and go up to 13.5%, which makes the proposed algorithm very suitable for future video coding solutions focusing on high fidelity applications.

In our previous work, we introduced spatially varying transforms (SVT) for video coding, where the location of the transform block within the macroblock is not fixed but varying. SVT has lower decoding complexity compared to standard methods as only a portion of the prediction error needs to be decoded. However, the encoding complexity of SVT can be relatively high because of the need to perform rate distortion optimization (RDO) for each candidate location parameter (LP). In this work, we propose a low complexity algorithm operating on macroblock and block level to reduce the encoding complexity of SVT. The proposed low complexity algorithm includes selection of available candidate LP based on motion difference and a hierarchical search algorithm. Experimental results show that the proposed low complexity algorithm can reduce around 80% of the candidate LP tested in RDO with only marginal penalty in coding efficiency.

In this paper, a novel algorithm called spatially varying transform (SVT) is proposed to improve the coding efficiency of video coders. SVT enables video coders to vary the position of the transform block, unlike state-of-art video codecs where the position of the transform block is fixed. In addition to changing the position of the transform block, the size of the transform can also be varied within the SVT framework, to better localize the prediction error so that the underlying correlations are better exploited. It is shown in this paper that by varying the position of the transform block and its size, characteristics of prediction error are better localized, and the coding efficiency is thus improved. The proposed algorithm is implemented and studied in the H.264/AVC framework. We show that the proposed algorithm achieves 5.85% bitrate reduction compared to H.264/AVC on average over a wide range of test set. Gains become more significant at medium to high bitrates for most tested sequences and the bitrate reduction may reach 13.50%, which makes the proposed algorithm very suitable for future video coding solutions focusing on high fidelity video applications. The gain in coding efficiency is achieved with a similar decoding complexity which makes the proposed algorithm easy to be incorporated in video codecs. However, the encoding complexity of SVT can be relatively high because of the need to perform a number of rate distortion optimization (RDO) steps to select the best location parameter (LP), which indicates the position of the transform. In this paper, a novel low complexity algorithm is also proposed, operating on a macroblock and a block level, to reduce the encoding complexity of SVT. Experimental results show that the proposed low complexity algorithm can reduce the number of LPs to be tested in RDO by about 80% with only a marginal penalty in the coding efficiency.

Adaptive prediction error coding (APEC) in spatial and frequency domain is proposed and significantly improves the coding efficiency of video coders. However, this approach comes with the expense of increased encoding complexity because of the need to perform rate distortion optimization (RDO) for each block, to decide whether the block is coded in spatial or frequency domain. In this work, we propose novel fast mode decision algorithm operating on macroblock and block level to reduce the encoding complexity of APEC. The proposed algorithm estimates the edge orientation of the residual block using spatial domain filtering, and based on the estimated edge orientation, reduces the number of candidates to be checked in RDO. Furthermore, proposed method includes an early termination step that stops searching the best candidate based on the difference between spatial and frequency domain coding. Experimental results show that the proposed fast mode decision algorithm can achieve averagely 90.7% encoding time reduction of APEC with only about 0.06 dB loss in coding efficiency.

We propose an algorithm for designing nonuniform oversam-pled filterbanks with arbitray delay. The filterbank has uni-form sections obtained by generalized DFT modulation; be-tween the uniform sections, there are transition filters. There is no a priori ...

A technique is developed for the design of analysis filters in an M-channel maximally decimated, perfect reconstruction, finite-impulse-response quadrature mirror filter (FIR QMF) bank that has a lossless polyphase-component matrix E (z). The aim is to ...

A very efficient technique to drastically reduce the number of multipliers and adders in narrow transition-band linear-phase finite-impulse response digital filters is to use the one-stage or multistage frequency-response masking (FRM) approach, which has been originally introduced by Lim and further improved by Lim and Lian. In these original synthesis techniques, the subfilters in the overall implementation are separately designed. As shown earlier by the authors of this contribution together with Johansson, the arithmetic complexity in one-stage FRM filter designs can be considerably reduced by using the following two-step technique for simultaneously optimizing all the subfilters. First, a suboptimal solution is found by using a simple design scheme. Second, this solution is used as a start-up solution for further optimization, which is carried out with the aid of an efficient nonlinear optimization algorithm. This paper exploits this approach to synthesizing multistage FRM filters. An example taken from the literature illustrates that both the number of multipliers and the number of adders for the resulting optimized multistage FRM filters are approximately 70 percent compared with those of the filters synthesized using the original multistage FRM filter design schemes. Additional examples are included in order to show the benefits provided by the proposed synthesis scheme over other recently published design techniques, in terms of an improved performance of Some parts of this paper were presented at

An efficient coefficient quantization scheme is described for minimizing the cost for implementing the fixed parallel linear-phase finite-impulse response (FIR) filters in the modified Farrow structure introduced by Vesma and Saramäki for generating FIR filters with an adjustable fractional delay. The implementation costs under consideration are the minimum number of adders and subtracters when implementing these parallel subfilters as very-largescale integration (VLSI) circuit. Two implementation costs are under consideration to meet the given criteria. In the first case, all the coefficient values are implemented independently of each others as a few signed powers-of-two terms, whereas in the second case, the common subexpressions within all the coefficient values included in the overall implementation are properly shared in order to reduce the overall implementation cost even further. The optimum finiteprecision solution is found in four steps. First, the number of filters and their (common odd) order are determined such that the given criteria are sufficiently exceeded in order to allow some coefficient quantization errors. Second, those coefficient values of the subfilters having a negligible effect on the overall system performance are fixed to be zero-valued. In addition, the experimentally observed attractive connections between the coefficient values of the subfilters, after setting some coefficient values equal to zero, are utilized in order to reduce both the implementation cost and the parameters to be optimized even more. Third, constrained nonlinear optimization is applied to determining for the remaining infinite-precision coefficients a parameter space that includes the feasible space where the given criteria are met. The fourth step involves finding in this space the desired finite-precision coefficient values for minimizing the given implementation costs to meet the stated overall criteria. Several examples are included illustrating the efficiency of the proposed synthesis scheme.

We propose an efficient algorithm for designing the prototype filters of oversampled, near perfect reconstruction (NPR), GDFT modulated, biorthogonal filterbanks with arbitrary delay. Given the analysis prototype, we show that the minimization of the stopband energy of the synthesis prototype, subject to NPR constraints on the frequency response of the distortion transfer function, can be expressed as a convex optimization problem. Our algorithm consists of initialization with the prototype of an orthogonal filterbank and then successive optimization of the synthesis and analysis prototypes. We compare our algorithm with previous methods and give several design examples.

This paper describes an efficient algorithm for designing lattice wave digital (LWD) filters (parallel connections of two all-pass filters) with short coefficient wordlength. The coefficient optimization is performed using the following three steps. First, an initial infinite-precision filter is designed such that it exceeds the given criteria in order to provide some tolerance for coefficient quantization. Second, a nonlinear optimization algorithm is used for determining a parameter space of the infinite-precision coefficients including the feasible space where the filter meets the given criteria. The third step involves finding the filter parameters in this space so that the resulting filter meets the given criteria with the simplest coefficient representation forms.

This paper proposes a computational, very efficient, approach for designing a novel family of -channel maximally decimated nearly perfect-reconstruction cosine-modulated transmultiplexers. This approach is referred to as the generalized windowing method for transmultiplexers because after knowing the transmission channel a proper weighted sum of the inter-channel and inter-symbol interferences can be properly taken into account in the optimization of the window function, unlike in other existing windowing techniques. The proposed approach has also the following two advantages. First, independent of the number of subchannels and the common order of the subchannel filters, the number of unknowns is only four. Second, the overall optimization procedure is made considerably fast by estimating the above-mentioned sum in terms of two novel measures, namely, the signal to inter-symbol and the signal to inter-channel interferences, which are very easy to evaluate. Furthermore, when the transmission channel is not considered in the design, a table is provided, which contains the parameters for designing the prototype filter directly by using the windowing method without any time-consuming optimization. When comparing the resulting transmultiplexers with the corresponding perfect-reconstruction designs (the same number of subchannels and same prototype filter order), the levels of interferences are practically the same. However, when the system is affected by a strong narrowband interference, the proposed transmultiplexers outperform their PR counterparts. Design examples are included illustrating the efficiency of the proposed design approach over other existing techniques based on the use of the windowing method.