E-FineR is a training-free, fully automated framework for vocabulary-free fine-grained visual recognition. This repository accompanies the research paper: Vocabulary-free Fine-grained Visual ...