Paper ID: 2303.09269

ELFIS: Expert Learning for Fine-grained Image Recognition Using Subsets

Pablo Villacorta, Jesús M. Rodríguez-de-Vera, Marc Bolaños, Ignacio Sarasúa, Bhalaji Nagarajan, Petia Radeva

Fine-Grained Visual Recognition (FGVR) tackles the problem of distinguishing highly similar categories. One of the main approaches to FGVR, namely subset learning, tries to leverage information from existing class taxonomies to improve the performance of deep neural networks. However, these methods rely on the existence of handcrafted hierarchies that are not necessarily optimal for the models. In this paper, we propose ELFIS, an expert learning framework for FGVR that clusters categories of the dataset into meta-categories using both dataset-inherent lexical and model-specific information. A set of neural networks-based experts are trained focusing on the meta-categories and are integrated into a multi-task framework. Extensive experimentation shows improvements in the SoTA FGVR benchmarks of up to +1.3% of accuracy using both CNNs and transformer-based networks. Overall, the obtained results evidence that ELFIS can be applied on top of any classification model, enabling the obtention of SoTA results. The source code will be made public soon.

Submitted: Mar 16, 2023