Throughput Scaling Of Convolution For Error-Tolerant Multimedia Applications

TL;DR

This paper introduces a method to increase convolution throughput in multimedia processing by allowing controlled imprecision, enabling faster computation with minimal impact on accuracy for error-tolerant applications.

Contribution

It proposes two novel throughput scaling techniques using scalar quantization and floating-point packing, tailored for error-tolerant multimedia applications.

Findings

Achieves up to 175% throughput increase over full-precision convolution.

Maintains virtually no accuracy loss in multimedia tasks.

Adjusts throughput and distortion based on input data statistics.

Abstract

Convolution and cross-correlation are the basis of filtering and pattern or template matching in multimedia signal processing. We propose two throughput scaling options for any one-dimensional convolution kernel in programmable processors by adjusting the imprecision (distortion) of computation. Our approach is based on scalar quantization, followed by two forms of tight packing in floating-point (one of which is proposed in this paper) that allow for concurrent calculation of multiple results. We illustrate how our approach can operate as an optional pre- and post-processing layer for off-the-shelf optimized convolution routines. This is useful for multimedia applications that are tolerant to processing imprecision and for cases where the input signals are inherently noisy (error tolerant multimedia applications). Indicative experimental results with a digital music matching system and an MPEG-7 audio descriptor system demonstrate that the proposed approach offers up to 175% increase in processing throughput against optimized (full-precision) convolution with virtually no effect in the accuracy of the results. Based on marginal statistics of the input data, it is also shown how the throughput and distortion can be adjusted per input block of samples under constraints on the signal-to-noise ratio against the full-precision convolution.