Park, Soonyong Park, Sung-Kee Hebert, Martial 2024-01-20T10:30:35Z 2024-01-20T10:30:35Z 2021-09-04 2014-03 0162-8828 https://pubs.kist.re.kr/handle/201004/127065 This paper presents a fast and efficient computational approach to higher order spectral graph matching. Exploiting the redundancy in a tensor representing the affinity between feature points, we approximate the affinity tensor with the linear combination of Kronecker products between bases and index tensors. The bases and index tensors are highly compressed representations of the approximated affinity tensor, requiring much smaller memory than in previous methods, which store the full affinity tensor. We compute the principal eigenvector of the approximated affinity tensor using the small bases and index tensors without explicitly storing the approximated tensor. To compensate for the loss of matching accuracy by the approximation, we also adopt and incorporate a marginalization scheme that maps a higher order tensor to matrix as well as a one-to-one mapping constraint into the eigenvector computation process. The experimental results show that the proposed method is faster and requires smaller memory than the existing methods with little or no loss of accuracy. English IEEE COMPUTER SOC ALGORITHM Fast and Scalable Approximate Spectral Matching for Higher Order Graph Matching Article 10.1109/TPAMI.2013.157 1 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, v.36, no.3, pp.479 - 492 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 36 3 479 492 scie scopus 000331450100007 2-s2.0-84894549705 Computer Science, Artificial Intelligence Engineering, Electrical & Electronic Computer Science Engineering Article ALGORITHM Higher order graph matching spectral relaxation approximation algorithm