Dependence Maximizing Temporal Alignment via Squared-Loss Mutual Information

TL;DR

This paper introduces LSDTW, a novel temporal alignment method that maximizes dependency between sequences using squared-loss mutual information, effective for sequences with different lengths, dimensions, and non-linearities.

Contribution

The paper presents LSDTW, a new information-theoretic approach for temporal alignment that is computationally efficient and adaptable to various sequence complexities.

Findings

LSDTW effectively aligns sequences with different lengths and dimensions.

The method outperforms traditional techniques on synthetic and real-world datasets.

Cross-validation optimizes model parameters for improved accuracy.

Abstract

The goal of temporal alignment is to establish time correspondence between two sequences, which has many applications in a variety of areas such as speech processing, bioinformatics, computer vision, and computer graphics. In this paper, we propose a novel temporal alignment method called least-squares dynamic time warping (LSDTW). LSDTW finds an alignment that maximizes statistical dependency between sequences, measured by a squared-loss variant of mutual information. The benefit of this novel information-theoretic formulation is that LSDTW can align sequences with different lengths, different dimensionality, high non-linearity, and non-Gaussianity in a computationally efficient manner. In addition, model parameters such as an initial alignment matrix can be systematically optimized by cross-validation. We demonstrate the usefulness of LSDTW through experiments on synthetic and real-world Kinect action recognition datasets.