CVPR MetaFood Workshop [CVPR2024]

We are happy to share that one of our papers "LOFI: LOng-tailed FIne-Grained Network for Food Recognition" has been accepted as an oral to the MetaFood Workshop at CVPR2024. The paper identifies the main challenges of visual food recognition (e.g. fine-grained and long-tailed distributions) and proposes LOFI, a new tailored method that successfully tackles them. LOFI achieves state-of-the-art results in a variety of datasets, showing its versatility and generalization ability. Food recognition tasks are key to many healthcare and industrial applications, such as food intake monitoring and food production chains. Thus, the improvements introduced by LOFI can directly benefit real-world users daily.Paper work done by Prof. Petia I. Radeva, Jesús Molina Rodríguez de Vera, Imanol G. Estepa, Marc Bolaños Solà, and Bhalaji Nagarajan.