Hypergraph matching via game-theoretic hypergraph clustering
作者：
Highlights：
• We formulate hypergraph matching of visual features as hypergraph clustering of candidate matches.
• We find it is costly to obtain a large number of matches with one single ESS group and discuss the reasons.
• We propose to extract an appropriate number of clusters to increase the number of matches efficiently.
• Experiments show that our algorithm generates a large number of matches with a high matching accuracy efficiently.
摘要
•We formulate hypergraph matching of visual features as hypergraph clustering of candidate matches.•We find it is costly to obtain a large number of matches with one single ESS group and discuss the reasons.•We propose to extract an appropriate number of clusters to increase the number of matches efficiently.•Experiments show that our algorithm generates a large number of matches with a high matching accuracy efficiently.
论文关键词：Feature matching,Hypergraph matching,Game-theoretic,Hypergraph clustering
论文评审过程：Received 2 June 2021, Revised 5 December 2021, Accepted 7 January 2022, Available online 10 January 2022, Version of Record 14 January 2022.
论文官网地址：https://doi.org/10.1016/j.patcog.2022.108526
