Tensor-based approach to accelerate deformable part models

TL;DR

This paper introduces a tensor-based method to significantly speed up deformable part models by reducing convolution operations, achieving up to 4.5 times faster processing with minimal accuracy loss.

Contribution

It presents a novel tensor decomposition approach for filters that accelerates DPMs, balancing performance and accuracy more effectively than previous methods.

Findings

Achieved up to 4.5x reduction in convolutions

Maintained similar accuracy to original DPMs

Validated on Pascal VOC dataset

Abstract

This article provides next step towards solving speed bottleneck of any system that intensively uses convolutions operations (e.g. CNN). Method described in the article is applied on deformable part models (DPM) algorithm. Method described here is based on multidimensional tensors and provides efficient tradeoff between DPM performance and accuracy. Experiments on various databases, including Pascal VOC, show that the proposed method allows decreasing a number of convolutions up to 4.5 times compared with DPM v.5, while maintaining similar accuracy. If insignificant accuracy degradation is allowable, higher computational gain can be achieved. The method consists of filters tensor decomposition and convolutions shortening using the decomposed filter. Mathematical overview of the proposed method as well as simulation results are provided.