Regression-Based Elastic Metric Learning on Shape Spaces of Cell Curves

Abstract

We propose a metric learning paradigm, Regression-based Elastic Metric Learning (REML), which optimizes the elastic metric for geodesic regression on the manifold of discrete curves. Geodesic regression is most accurate when the chosen metric models the data trajectory close to a geodesic on the discrete curve manifold. When tested on cell shape trajectories, regression with REML’s learned metric has better predictive power than with the conventionally used square-root-velocity (SRV) metric. The code is publicly available here: https://github.com/bioshape-lab/dyn.

PDF NeurIPSW OpenReview Semantic Scholar

Cite

Text

Myers and Miolane. "Regression-Based Elastic Metric Learning on Shape Spaces of Cell Curves." NeurIPS 2022 Workshops: LMRL, 2022.

Markdown

[Myers and Miolane. "Regression-Based Elastic Metric Learning on Shape Spaces of Cell Curves." NeurIPS 2022 Workshops: LMRL, 2022.](https://mlanthology.org/neuripsw/2022/myers2022neuripsw-regressionbased/)

BibTeX

@inproceedings{myers2022neuripsw-regressionbased,
  title     = {{Regression-Based Elastic Metric Learning on Shape Spaces of Cell Curves}},
  author    = {Myers, Adele and Miolane, Nina},
  booktitle = {NeurIPS 2022 Workshops: LMRL},
  year      = {2022},
  url       = {https://mlanthology.org/neuripsw/2022/myers2022neuripsw-regressionbased/}
}