Note on the Construction of Structure Tensor
Josef Bigun, Fernado Alonso-Fernandez

TL;DR
This paper unifies two structure tensor constructions through Total Least Squares, simplifying interpretation, removing unnecessary correction terms, and broadening the scope to include various filter types like Gabor filters.
Contribution
It demonstrates that two seemingly different structure tensor methods can be reconciled via TLS, leading to simplified interpretation and greater flexibility in filter choice.
Findings
The correction term in Granlund and Knutsson 1995 is unnecessary when viewed through TLS.
The resulting tensor remains positive semi-definite without the correction.
Alternative filters like Gabor can be used for structure tensor construction.
Abstract
This note presents a theoretical discussion of two structure tensor constructions: one proposed by Bigun and Granlund 1987, and the other by Granlund and Knutsson 1995. At first glance, these approaches may appear quite different--the former is implemented by averaging outer products of gradient filter responses, while the latter constructs the tensor from weighted outer products of tune-in frequency vectors of quadrature filters. We argue that when both constructions are viewed through the common lens of Total Least Squares (TLS) line fitting to the power spectrum, they can be reconciled to a large extent, and additional benefits emerge. From this perspective, the correction term introduced in Granlund and Knutsson 1995 becomes unnecessary. Omitting it ensures that the resulting tensor remains positive semi-definite, thereby simplifying the interpretation of its eigenvalues.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
