Shape Morphing Evaluation with the Geodesic Deformation Shape Estimation

This project implements a shape morphing evaluation method using the Geodesic Closest Path estimation (GCP) criterion and a knot-detector inspired by Ghorbel & Ghorbel work.

Features

• Shape morphing using interpolation (replaceable with other planar morphing methods)
• GCP criterion to evaluate the closness of morph paths to the geodesic in the shape space.
• Procrustes-based alignment (can be replaced by pseudo-inverse registration algorithm)
• Clean visualization of results

How to Run

pip install -r requirements.txt
python main.py

Citation

Ghorbel, E., Ghorbel, F. Data augmentation based on shape space exploration for low-size datasets: application to 2D shape classification. Neural Comput & Applic 36, 10031–10054 (2024). https://doi.org/10.1007/s00521-024-09798-5