Fast Cubical Persistent Homology on 2D and 3D Images via Union-Find, Pruning, and Lookup Tables
Researchers have developed Flash Cubical, a new method for efficiently computing cubical persistence on 2D and 3D images using a V-filtration over $\mathbb{F}_2$. The approach leverages union-find with pruning for efficient persistence computation and employs a lookup table to pre-compute local information, thereby avoiding runtime calculations. This method is claimed to be the most efficient in terms of both time and memory costs for this specific type of computation and may generalize to other filtrations and complexes. AI