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

@article{Oron2018BestBuddiesST,
  title={Best-Buddies Similarity—Robust Template Matching Using Mutual Nearest Neighbors},
  author={Shaul Oron and Tali Dekel and Tianfan Xue and William T. Freeman and Shai Avidan},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year={2018},
  volume={40},
  pages={1799-1813}
}
  • Shaul Oron, Tali Dekel, S. Avidan
  • Published 6 September 2016
  • Computer Science
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
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… 
Best Buddies Registration for Point Clouds
TLDR
Several algorithms, collectively named Best Buddy Registration (BBR), are presented, where each algorithm consists of optimizing one of these loss functions with Adam gradient descent, inspired by the Best Buddies Similarity measure that counts the number of mutual nearest neighbors between two point sets.
Multi-scale Template Matching with Scalable Diversity Similarity in an Unconstrained Environment
TLDR
A novel multi-scale template matching method which is robust against both scaling and rotation in unconstrained environments and can significantly outperform state-of-the-art methods is proposed.
Weighted Smallest Deformation Similarity for NN-Based Template Matching
TLDR
A weighted smallest deformation similarity measure, which is robust to occlusions, background outliers, and complex deformations, and can be applied in rough positioning of surface mount technology components is proposed.
Illumination robust template matching based on nearest neighbor
TLDR
Based on the relative pixel values that do not change with illumination, an illumination-robust distance function is defined, which is shown to be a well-performing similarity measure against illumination changes, non-rigid geometric deformations, as well as background clutter, and occlusions.
Scale-Adaptive NN-Based Similarity for Robust Template Matching
TLDR
Experiment on the real-world scenario benchmarks and surface-mount component (SMC) positioning shows that the proposed method is robust to scale changes along with other challenging aspects and outperforms the state-of-the-arts using both color and deep features.
Template Matching with Deformable Diversity Similarity
TLDR
A novel measure for template matching named Deformable Diversity Similarity &#x2013, based on the diversity of feature matches between a target image window and the template, that is robust to complex deformations, significant background clutter, and occlusions is proposed.
A Highly Efficient and Robust Method For NNF-Based Template Matching
TLDR
This paper proposes a highly efficient and robust method for template matching that utilizes the inverse NNF by searching the nearest neighbor of template points, which is more robust than original NNF.
DeepBBS: Deep Best Buddies for Point Cloud Registration
TLDR
This work proposes DeepBBS, a novel method for learning a representation that takes into account the best buddy distance between points during training, which leads to an accurate registration for partial shapes and in unseen categories.
...
...

References

SHOWING 1-10 OF 50 REFERENCES
Best-Buddies Similarity for robust template matching
TLDR
This work proposes a novel method for template matching in unconstrained environments, based on counting the number of Best-Buddies Pairs—pairs of points in source and target sets, where each point is the nearest neighbor of the other.
Asymmetric Correlation: A Noise Robust Similarity Measure for Template Matching
TLDR
An efficient and noise robust template matching method based on asymmetric correlation (ASC), which is invariant to affine illumination changes and robust to extreme noise and extended to deal with partial occlusion and spatially varying light change.
Fast-Match: Fast Affine Template Matching
TLDR
Fast-Match is a fast algorithm for approximate template matching under 2D affine transformations that minimizes the Sum-of-Absolute-Differences (SAD) error measure and it is proved that they can be sampled using a density that depends on the smoothness of the image.
Image retrieval with reciprocal and shared nearest neighbors
TLDR
This paper extends previous work on k-reciprocal nearest neighbors to produce new measures that improve over the original primary metric and proposes measures defined on sets of shared nearest neighbors for reranking the shortlist.
Reliable feature matching across widely separated views
  • A. Baumberg
  • Computer Science
    Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662)
  • 2000
TLDR
A robust method for automatically matching features in images corresponding to the same physical point on an object seen from two arbitrary viewpoints that is optimised for a structure-from-motion application where it wishes to ignore unreliable matches at the expense of reducing the number of feature matches.
Hello neighbor: Accurate object retrieval with k-reciprocal nearest neighbors
TLDR
This paper introduces a simple yet effective method to improve visual word based image retrieval based on an analysis of the k-reciprocal nearest neighbor structure in the image space and demonstrates a significant improvement over standard bag-of-words retrieval.
Fast and high-performance template matching method
TLDR
A new template matching method that is robust to outliers and fast enough for real-time operation is proposed and Challenging applications, such as intra-category object detection, object tracking, and multimodal image matching are demonstrated.
Fast and robust template matching algorithm in noisy image
TLDR
This paper proposes fast and robust template matching algorithm that uses grey level index table and image rank technique that can find specific area under the given template query image with 30% Gaussian noise.
Robust Real-Time Pattern Matching Using Bayesian Sequential Hypothesis Testing
  • Ofir Pele, M. Werman
  • Computer Science
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2008
TLDR
A novel Bayesian framework for sequential hypothesis testing on finite populations is presented, and an optimal rejection/acceptance sampling algorithm is designed that quickly determines whether two images are similar with respect to a member of the Image Hamming Distance Family.
Fast algorithm for robust template matching with M-estimators
TLDR
A fast algorithm is proposed for speeding up the process of template matching that uses M-estimators for dealing with outliers and ensures finding the global minimum of the robust template matching problem in which a nondecreasing M-ESTimator serves as an error measure.
...
...