Learn More for Food Recognition via Progressive Self-Distillation

TL;DR

This paper introduces a Progressive Self-Distillation method for food recognition that improves the network's ability to identify detailed regions without relying on explicit localization, achieving state-of-the-art results.

Contribution

The proposed PSD method uses a teacher-student framework with progressive training to enhance food recognition accuracy without explicit region localization.

Findings

Achieves state-of-the-art performance on three food datasets.

Effectively mines discriminative regions without explicit localization.

Improves recognition accuracy through progressive self-distillation.

Abstract

Food recognition has a wide range of applications, such as health-aware recommendation and self-service restaurants. Most previous methods of food recognition firstly locate informative regions in some weakly-supervised manners and then aggregate their features. However, location errors of informative regions limit the effectiveness of these methods to some extent. Instead of locating multiple regions, we propose a Progressive Self-Distillation (PSD) method, which progressively enhances the ability of network to mine more details for food recognition. The training of PSD simultaneously contains multiple self-distillations, in which a teacher network and a student network share the same embedding network. Since the student network receives a modified image from its teacher network by masking some informative regions, the teacher network outputs stronger semantic representations than the student network. Guided by such teacher network with stronger semantics, the student network is encouraged to mine more useful regions from the modified image by enhancing its own ability. The ability of the teacher network is also enhanced with the shared embedding network. By using progressive training, the teacher network incrementally improves its ability to mine more discriminative regions. In inference phase, only the teacher network is used without the help of the student network. Extensive experiments on three datasets demonstrate the effectiveness of our proposed method and state-of-the-art performance.