Continuous dynamic time warping for translation-invariant curve alignment with applications to signature verification

Munich, Mario E.; Perona, Pietro

doi:10.1109/iccv.1999.791205

Cited by 117 publications

(66 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, in bioinformatics, it is well understood that functionally related genes will express themselves in similar ways, but possibly at different rates. Because of this, warping distances have been used for gene expression data mining (Aach & Church, 2001;Bar-Joseph et al, 2002), for tracking time series extracted from video (Gavrila & Davis, 1995), classifying handwritten text (Rath & Manmatha, 2002;Munich & Perona, 1999) and even fingerprint indexing (Kovács-Vajna, 2000). Finally, perhaps the most familiar example of easy classifiable patterns, which nevertheless cause problems for Lp metrics, are electrocardiograms.…”

Section: Related Workmentioning

confidence: 99%

Elastic Translation Invariant Matching of Trajectories

Vlachos¹,

Kollios

Gunopulos

2005

Mach Learn

View full text Add to dashboard Cite

Abstract.We investigate techniques for analysis and retrieval of object trajectories. We assume that a trajectory is a sequence of two or three dimensional points. Trajectory datasets are very common in environmental applications, mobility experiments, video surveillance and are especially important for the discovery of certain biological patterns. Such kind of data usually contain a great amount of noise, that makes all previously used metrics fail. Therefore, here we formalize non-metric similarity functions based on the Longest Common Subsequence (LCSS), which are very robust to noise and furthermore provide an intuitive notion of similarity between trajectories by giving more weight to the similar portions of the sequences. Stretching of sequences in time is allowed, as well as global translating of the sequences in space. Efficient approximate algorithms that compute these similarity measures are also provided. We compare these new methods to the widely used Euclidean and Dynamic Time Warping distance functions (for real and synthetic data) and show the superiority of our approach, especially under the strong presence of noise. We prove a weaker version of the triangle inequality and employ it in an indexing structure to answer nearest neighbor queries. Finally, we present experimental results that validate the accuracy and efficiency of our approach.

show abstract

Section: Related Workmentioning

confidence: 99%

Elastic Translation Invariant Matching of Trajectories

Vlachos¹,

Kollios

Gunopulos

2005

Mach Learn

View full text Add to dashboard Cite

show abstract

“…Angel [18] used analytical DP for smooth interpolation. Serra and Berthod [20] and Munich and Perona [21] used it for nonlinear alignment of one-dimensional patterns. Finally, Uchida et al [22] used it for object tracking.…”

Section: Related Workmentioning

confidence: 99%

Analytical Dynamic Programming Matching

Uchida

Hokahori

Feng

2012

Computer Vision – ECCV 2012. Workshops and Demonstrations

View full text Add to dashboard Cite

Abstract. In this paper, we show that the truly two-dimensional elastic image matching problem can be solved analytically using dynamic programming (DP) in polynomial time if the problem is formulated as a maximum a posteriori problem using Gaussian distributions for the likelihood and prior. After giving the derivation of the analytical DP matching algorithm, we evaluate its performance on handwritten character images containing various nonlinear deformations, and compare other elastic image matching methods.

show abstract

“…Vector quantization (VQ) [3] pattern recognition algorithm has also been tested [3] in this field. Dynamic time warping [1], [7] and hidden Markov model(HMM) [2] are two of the most widely used approaches followed by Vector quantization(VQ) In the Signature Verification Competition 2004 (SVC04), DTW and HMM based systems were shown to be the most competitive algorithms.…”

Section: Related Workmentioning

confidence: 99%

“…Signatures are composed of special characters and flourishes therefore most of the time they can be unreadable. Also intrapersonal variations and interpersonal differences make it necessary to analyze them as complete images and not as letters and words put together [1]. As signatures are the primary mechanism both for authentication and authorization in legal transactions, the need for research in efficient auto-mated solutions for signature recognition and verification has increased in recent years.…”

Section: Introductionmentioning

confidence: 99%