Why Walking the Dog Takes Time: Frechet Distance Has No Strongly Subquadratic Algorithms Unless SETH Fails

@article{Bringmann2014WhyWT,
  title={Why Walking the Dog Takes Time: Frechet Distance Has No Strongly Subquadratic Algorithms Unless SETH Fails},
  author={Karl Bringmann},
  journal={2014 IEEE 55th Annual Symposium on Foundations of Computer Science},
  year={2014},
  pages={661-670}
}
  • K. Bringmann
  • Published 5 April 2014
  • Computer Science, Mathematics
  • 2014 IEEE 55th Annual Symposium on Foundations of Computer Science
The Fréchet distance is a well-studied and very popular measure of similarity of two curves. Many variants and extensions have been studied since Alt and Godau introduced this measure to computational geometry in 1991. Their original algorithm to compute the Fréchet distance of two polygonal curves with n vertices has a runtime of O(n^2 log n). More than 20 years later, the state of the art algorithms for most variants still take time more than O(n2 / log n), but no matching lower bounds are… 
View on IEEE
arxiv.org
236 Citations
Highly Influential Citations
21
Background Citations
100
Methods Citations
37
Results Citations
8

236 Citations

Four Soviets Walk the Dog: Improved Bounds for Computing the Fréchet Distance

This work gives a randomized algorithm to compute the Fréchet distance between two polygonal curves in time and shows that there exists an algebraic decision tree for the decision problem of depth, for some varepsilon > 0, which reveals an intriguing new aspect of this well-studied problem.

When Lipschitz Walks Your Dog: Algorithm Engineering of the Discrete Fréchet Distance under Translation

The solution combines fast, but inexact tools from continuous optimization with exact, but expensive algorithms from computational geometry to obtain an exact decision algorithm for the Frechet distance under translation.

Approximating the (continuous) Fr\'echet distance

This work describes the first strongly subquadratic time algorithm with subexponential approximation ratio for approximately computing the Fréchet distance between two polygonal chains, and describes how to turn any approximate decision procedure for the FrÉchet distance into an approximate optimization algorithm whose approximation ratio is the same up to arbitrarily small constant factors.

APPROXIMABILITY OF THE DISCRETE FRÉCHET

A new conditional lower bound is presented showing that strongly subquadratic algorithms for the discrete Fréchet distance are unlikely to exist, even in the one-dimensional case and even if the solution may be approximated up to a factor of 1.399.

Approximability of the discrete Fréchet distance

A new conditional lower bound is presented showing that strongly subquadratic algorithms for the discrete Frechet distance are unlikely to exist, even in the one-dimensional case and even if the solution may be approximated up to a factor of 1.399.

The k-Fréchet distance

It is shown that deciding this distance measure turns out to be NP-complete, which is interesting since both (weak) Fréchet and Hausdorff distance are computable in polynomial time.

Improved Approximation for Fréchet Distance on c-Packed Curves Matching Conditional Lower Bounds

This paper presents an improved algorithm with time complexity Open image in new window that improves upon the algorithm by Driemel et al. and matches the conditional lower bound (up to lower order factors of the form \(n^{o(1)}\)).

On Computing the k-Shortcut Fréchet Distance

A complexity analysis for the shortcut Fréchet distance, where one is allowed to take shortcuts along one of the curves, similar to the edit distance for sequences, and shows that efficient approximate decider algorithms are possible, even when k is large.

Discrete Fréchet Distance under Translation

This article provides evidence that constructing the arrangement of size Õ(N4) is necessary in the worst case by proving a conditional lower bound of n4 - o(1) on the running time for the discrete Fréchet distance under translation, assuming the Strong Exponential Time Hypothesis.

Computing the Fréchet Distance between Real-Valued Surfaces

This paper measures the distance between terrains based solely on the height function, and shows that in this case computing the Frechet distance between f and g is in NP, and defines an intermediate distance, between contour trees, which is shown to be NP-complete to compute.
...

References

SHOWING 1-10 OF 39 REFERENCES

Four Soviets Walk the Dog - with an Application to Alt's Conjecture

A randomized algorithm to compute the Frechet distance between two polygonal curves in time [EQUATION] on a pointer machine and in time O(n2(log log n)2) on a word RAM is given and there is evidence that the decision problem may not be 3SUM-hard after all.

Jaywalking Your Dog: Computing the Fréchet Distance with Shortcuts

The notion of a more robust Frechet distance, where one is allowed to shortcut between vertices of one of the curves, is introduced, which is a natural approach for handling noise, in particular batched outliers.

How Difficult is it to Walk the Dog ?

An Ω(n log n) lower bound for the decision problem in the algebraic computation tree model allowing arithmetic operations and tests is proved for two polygonal curves in the plane.

Fréchet distance with speed limits

Can We Compute the Similarity between Surfaces?

It follows that the decision problem, whether the Fréchet distance of two given surfaces lies below a specified value, is recursively enumerable, and it is shown that a relaxed version of the FrÉchet distance, the weak Fréchechet distance can be computed in polynomial time.

Computing the Discrete Fréchet Distance in Subquadratic Time

This work presents the first subquadratic algorithm for computing the discrete Frechet distance between two sequences of points in the plane, and uses the geometry of the problem in a subtle way to encode legal positions of the frogs as states of a finite automaton.

Exact algorithms for partial curve matching via the Fréchet distance

This paper presents the first exact algorithm that runs in polynomial time to compute fδ(P, Q), the partial Frechet similarity between P and Q, under the L1 and L∞ norms.

The Discrete Fréchet Distance with Shortcuts via Approximate Distance Counting and Selection

A variant of this technique, called the discrete Fréchet distance with shortcuts, which captures the similarity between (sampled) curves in the presence of outliers, and can apply to optimization problems for which the decision procedure is very fast but standard techniques like parametric search make the optimization algorithm substantially slower.

Comparison of Distance Measures for Planar Curves

It is shown that for closed convex curves both distance measures are the same and are within a constant factor of each other for so-called κ-straight curves, i.e., curves where the arc length between any two points on the curve is at most a constant κ times their Euclidean distance.

The fréchet distance revisited and extended

Given two simplicial complexes in Rd and start and end vertices in each complex, we show how to compute curves (in each complex) between these vertices, such that the weak Fréchet distance between…