Hypergraph-guided Intra- and Inter-category Relation Modeling for Fine-grained Visual Recognition

Lu Chen, Qiangchang Wang, Zhaohui Li, Yilong Yin

Abstract

Fine-grained Visual Recognition (FGVR) aims to distinguish objects within similar subcategories. Humans adeptly perform this challenging task by leveraging both intra-category distinctiveness and inter-category similarity. However, previous methods failed to combine these two complementary dimensions and mine the intrinsic interrelationship among various semantic features. To address the above limitations, we propose HI2R, a Hypergraph-guided Intra- and Inter-category Relation Modeling approach, which simultaneously extracts the intra-category structural information and inter-category relation information for more precise reasoning. Specifically, we exploit a Hypergraph-guided Structure Learning (HSL) module, which employs hypergraphs to capture high-order structural relations, transcending traditional graph-based methods that are limited to pairwise linkages. This advancement allows the model to adapt to significant intra-category variations. Additionally, we propose an Inter-category Relation Perception (IRP) module to improve feature discrimination across categories by extracting and analyzing semantic relations among them. Our objective is to alleviate the robustness issue associated with exclusive reliance on intra-category discriminative features. Furthermore, a random semantic consistency loss is introduced to direct the model's attention to commonly overlooked yet distinctive regions, which indirectly enhances the representation ability of both HSL and IRP modules. Both qualitative and quantitative results demonstrate the effectiveness and usefulness of our proposed HI2R model.
BibTeX
@inproceedings{chen2024hypergraph,
  author = {Lu Chen and Qiangchang Wang and Zhaohui Li and Yilong Yin},
  title = {{Hypergraph-guided Intra- and Inter-category Relation Modeling for Fine-grained Visual Recognition}},
  booktitle = {ACM Multimedia},
  year = {2024},
  doi = {10.1145/3664647.3680589},
}

Loading related content...

Figure