Best-Buddies Similarity - Robust Template Matching Using Mutual Nearest Neighbors

Shaul Oron*, Tali Dekel, Tianfan Xue, William T. Freeman, Shai Avidan

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


We propose a novel method for template matching in unconstrained environments. Its essence is the Best-Buddies Similarity (BBS), a useful, robust, and parameter-free similarity measure between two sets of points. BBS is based on counting the number of Best-Buddies Pairs (BBPs) - pairs of points in source and target sets that are mutual nearest neighbours, i.e., each point is the nearest neighbour of the other. BBS has several key features that make it robust against complex geometric deformations and high levels of outliers, such as those arising from background clutter and occlusions. We study these properties, provide a statistical analysis that justifies them, and demonstrate the consistent success of BBS on a challenging real-world dataset while using different types of features.

Original languageEnglish
Pages (from-to)1799-1813
Number of pages15
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Issue number8
StatePublished - 1 Aug 2018


  • Best buddies
  • mutual nearest neighbors
  • non-rigid matching
  • point set similarity
  • template matching


Dive into the research topics of 'Best-Buddies Similarity - Robust Template Matching Using Mutual Nearest Neighbors'. Together they form a unique fingerprint.

Cite this