Multiscale Manifold Warping

Many real-world applications require aligning two temporal sequences, including bioinformatics, handwriting recognition, activity recognition, and human-robot coordination. Dynamic Time Warping (DTW) is a popular alignment method, but can fail on high-dimensional real-world data where the dimensions of aligned sequences are often unequal. In this paper, we show that exploiting the multiscale manifold latent structure of real-world data can yield improved alignment. We introduce a novel framework called Warping on Wavelets (WOW) that integrates DTW with a a multi-scale manifold learning framework called Diffusion Wavelets. We present a theoretical analysis of the WOW family of algorithms and show that it outperforms previous state of the art methods, such as canonical time warping (CTW) and manifold warping, on several real-world datasets.

[1]  Chang Wang,et al.  Manifold Alignment Preserving Global Geometry , 2013, IJCAI.

[2]  I. Rojas,et al.  Estimation of COVID-19 dynamics in the different states of the United States using Time-Series Clustering , 2020, medRxiv.

[3]  Sameer A. Nene,et al.  Columbia Object Image Library (COIL100) , 1996 .

[4]  René Vidal,et al.  A closed form solution to robust subspace estimation and clustering , 2011, CVPR 2011.

[5]  Jessica K. Hodgins,et al.  Aligned Cluster Analysis for temporal segmentation of human motion , 2008, 2008 8th IEEE International Conference on Automatic Face & Gesture Recognition.

[6]  Chang Wang,et al.  Multiscale Manifold Learning , 2013, AAAI.

[7]  Davide Anguita,et al.  Human Activity Recognition on Smartphones with Awareness of Basic Activities and Postural Transitions , 2014, ICANN.

[8]  Mikhail Belkin,et al.  Laplacian Eigenmaps and Spectral Techniques for Embedding and Clustering , 2001, NIPS.

[9]  S. Chiba,et al.  Dynamic programming algorithm optimization for spoken word recognition , 1978 .

[10]  Xiaofei He,et al.  Locality Preserving Projections , 2003, NIPS.

[11]  Jessica K. Hodgins,et al.  Guide to the Carnegie Mellon University Multimodal Activity (CMU-MMAC) Database , 2008 .

[12]  S T Roweis,et al.  Nonlinear dimensionality reduction by locally linear embedding. , 2000, Science.

[13]  S. Mallat.  A wavelet tour of signal processing , 1998 .

[14]  Daniel D. Lee,et al.  Semisupervised alignment of manifolds , 2005, AISTATS.

[15]  Chang Wang,et al.  A General Framework for Manifold Alignment , 2009, AAAI Fall Symposium: Manifold Learning and Its Applications.

[16]  Patrick Pérez,et al.  Cross-View Action Recognition from Temporal Self-similarities , 2008, ECCV.

[17]  Sridhar Mahadevan,et al.  Manifold Warping: Manifold Alignment over Time , 2012, AAAI.

[18]  T. W. Anderson,et al.  An Introduction to Multivariate Statistical Analysis , 1959 .

[19]  R. Coifman,et al.  Diffusion Wavelets , 2004 .

[20]  U. Feige,et al.  Spectral Graph Theory , 2015 .

[21]  Fernando De la Torre,et al.  Canonical Time Warping for Alignment of Human Behavior , 2009, NIPS.

[22]  Sridhar Mahadevan,et al.  Aligning Mixed Manifolds , 2015, AAAI.

[23]  Yunqian Ma,et al.  Manifold Learning Theory and Applications , 2011 .