Background Splitting: Finding Rare Classes in a Sea of Background

TL;DR

This paper introduces Background Splitting, a method that leverages pre-trained models to better identify rare classes in datasets dominated by background images, significantly improving classification accuracy.

Contribution

The paper proposes an automatic background splitting technique using a pre-trained model and an auxiliary loss, enhancing rare class detection in highly imbalanced datasets without manual labels.

Findings

Achieves 42.3 mAP improvement over baseline in extreme background imbalance.

Outperforms state-of-the-art methods by 8.3 mAP on modified iNaturalist dataset.

Effectively reduces overfitting to background in rare class detection.

Abstract

We focus on the real-world problem of training accurate deep models for image classification of a small number of rare categories. In these scenarios, almost all images belong to the background category in the dataset (>95% of the dataset is background). We demonstrate that both standard fine-tuning approaches and state-of-the-art approaches for training on imbalanced datasets do not produce accurate deep models in the presence of this extreme imbalance. Our key observation is that the extreme imbalance due to the background category can be drastically reduced by leveraging visual knowledge from an existing pre-trained model. Specifically, the background category is "split" into smaller and more coherent pseudo-categories during training using a pre-trained model. We incorporate background splitting into an image classification model by adding an auxiliary loss that learns to mimic the predictions of the existing, pre-trained image classification model. Note that this process is automatic and requires no additional manual labels. The auxiliary loss regularizes the feature representation of the shared network trunk by requiring it to discriminate between previously homogeneous background instances and reduces overfitting to the small number of rare category positives. We also show that BG splitting can be combined with other background imbalance methods to further improve performance. We evaluate our method on a modified version of the iNaturalist dataset where only a small subset of rare category labels are available during training (all other images are labeled as background). By jointly learning to recognize ImageNet categories and selected iNaturalist categories, our approach yields performance that is 42.3 mAP points higher than a fine-tuning baseline when 99.98% of the data is background, and 8.3 mAP points higher than SotA baselines when 98.30% of the data is background.