Compositional Attribute Imbalance in Vision Datasets
This work addresses attribute imbalance for image classification, offering a scalable solution for long-tail tasks, but it is incremental as it builds on existing data augmentation techniques.
The paper tackled the problem of visual attribute imbalance in image classification by analyzing single-attribute and compositional attribute imbalances and proposing a sampling adjustment strategy integrated with data augmentation. The method improved robustness and fairness on benchmark datasets, effectively mitigating attribute imbalance.
Visual attribute imbalance is a common yet underexplored issue in image classification, significantly impacting model performance and generalization. In this work, we first define the first-level and second-level attributes of images and then introduce a CLIP-based framework to construct a visual attribute dictionary, enabling automatic evaluation of image attributes. By systematically analyzing both single-attribute imbalance and compositional attribute imbalance, we reveal how the rarity of attributes affects model performance. To tackle these challenges, we propose adjusting the sampling probability of samples based on the rarity of their compositional attributes. This strategy is further integrated with various data augmentation techniques (such as CutMix, Fmix, and SaliencyMix) to enhance the model's ability to represent rare attributes. Extensive experiments on benchmark datasets demonstrate that our method effectively mitigates attribute imbalance, thereby improving the robustness and fairness of deep neural networks. Our research highlights the importance of modeling visual attribute distributions and provides a scalable solution for long-tail image classification tasks.