Comparing Trajectories on the Size and Shape Space
Valerio Varano, Stefano Gabriele, Luciano Teresi, Ian Dryden, Paolo, Emilio Puddu, Concetta Torromeo, Paolo Piras

TL;DR
This paper emphasizes the importance of proper shape representation in trajectory analysis and introduces a novel Direct Transport method that improves deformation transport accuracy, especially in affine cases.
Contribution
It proposes a new Direct Transport procedure for better deformation transport in shape trajectory analysis, addressing limitations of existing methods.
Findings
Direct Transport perfectly transports deformation in affine cases.
Existing methods do not preserve deformation in affine cases.
Direct Transport outperforms existing tools in non-affine deformation approximation.
Abstract
In this paper we show that trajectory shape analysis should be performed only after obtaining a proper representation before applying ordination methods. In fact, studying the shape of a trajectory means studying how the deformation changes along each path irrespectively of the actual shape to which these deformations apply. The independence of the deformation from the shape to which it is applied is critical: it implies that any shape variation between individuals at the beginning of each trajectories must be completely filtered out. A Parallel Transport, that can be based on various connection types, is necessary to perform such kind of shape data centering. The Levi Civita connection can also be used to transport a deformation. We demonstrate that this procedure does not preserve deformation even in the ane case. We propose a novel procedure called Direct Transport able to perfectly…
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.
Taxonomy
TopicsMorphological variations and asymmetry · Image Processing and 3D Reconstruction · Image Retrieval and Classification Techniques
