Hi folks, I signal this contribution aimed at exploring some options offered by the use of deformation strain tensors in shape analysis (in the Size and Shape Space) beyond visualization.
https://www.mdpi.com/2227-7390/14/1/12 All simulated data (precious 2D data generated by assigning a-priori known local deformation tensors integrated over a compatible domain), experimental data (3D ontogenetic data of 4 primate species including Homo sapiens) , R scripts and functions (embedded in an undocumented R package that must be installed as source; scripts ar fully commented, however) are available to ensure full reproducibility and hopefully future researches made by others. The application to the study of longitudinal trajectories as an alternative to Parallel Transport methods is faced in detail. Of course limitations, warnings and caveats linked to some arbitrary choices implied by these procedures, as well as comparisons with previous attempts, are explicitly discussed. TPS, Quadratic Trend and Cubic Trend interpolants are compared as well as discrete computation of tensors on finite elements. Very far to be conclusive, we hope this can stimulate your interest about these alternatives. Thanks for your attention All the best Paolo -- You received this message because you are subscribed to the Google Groups "Morphmet" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/morphmet2/CAAdPgfMU2qgPNNTw%3DOWXXbF3idL6NR%2B1Yh4zR-%3DuX3bB2_4pUw%40mail.gmail.com.
