Skip to main content

The PatchMatch randomized matching algorithm for image manipulation

Author(s): Barnes, Connelly; Goldman, Dan B; Shechtman, Eli; Finkelstein, Adam

To refer to this page use:
Abstract: This paper presents a new randomized algorithm for quickly finding approximate nearest neighbor matches between image patches. Our algorithm offers substantial performance improvements over the previous state of the art (20--100×), enabling its use in new interactive image editing tools, computer vision, and video applications. Previously, the cost of computing such matches for an entire image had eluded efforts to provide interactive performance. The key insight driving our algorithm is that the elements of our search domain---patches of image pixels---are correlated, and thus the search strategy takes advantage of these statistics. Our algorithm uses two principles: first, that good patch matches can be found via random sampling, and second, that natural coherence in the imagery allows us to propagate such matches quickly to surrounding areas. Our simple algorithm allows finding a single nearest neighbor match across translations only, whereas our general algorithm additionally allows matching of k-nearest neighbors, across all rotations and scales, and matching arbitrary descriptors. This one simple algorithm forms the basis for a variety of applications including image retargeting, completion, reshuffling, object detection, digital forgery detection, and video summarization.
Publication Date: Nov-2011
Citation: Barnes, Connelly, Dan B. Goldman, Eli Shechtman, and Adam Finkelstein. "The patchmatch randomized matching algorithm for image manipulation." Communications of the ACM 54, no. 11 (2011): pp. 103-110. doi:10.1145/2018396.2018421
DOI: 10.1145/2018396.2018421
ISSN: 0001-0782
EISSN: 1557-7317
Pages: 103 - 110
Type of Material: Journal Article
Journal/Proceeding Title: Communications of the ACM
Version: Author's manuscript

Items in OAR@Princeton are protected by copyright, with all rights reserved, unless otherwise indicated.