AnyLoc: Towards Universal Visual Place Recognition

摘要

A universal visual place recognition approach using foundation model features for robust place recognition.