Text-Guided Mixup Towards Long-Tailed Image Categorization

TL;DR

This paper introduces a novel text-guided mixup method leveraging vision-language models like CLIP to improve long-tailed image classification by utilizing semantic relations from textual information, showing promising empirical results.

Contribution

It proposes a new text-guided mixup technique that uses pre-trained vision-language models to address long-tailed class distributions in image categorization.

Findings

Effective in long-tailed benchmarks

Leverages semantic relations from text

Theoretical guarantees provided

Abstract

In many real-world applications, the frequency distribution of class labels for training data can exhibit a long-tailed distribution, which challenges traditional approaches of training deep neural networks that require heavy amounts of balanced data. Gathering and labeling data to balance out the class label distribution can be both costly and time-consuming. Many existing solutions that enable ensemble learning, re-balancing strategies, or fine-tuning applied to deep neural networks are limited by the inert problem of few class samples across a subset of classes. Recently, vision-language models like CLIP have been observed as effective solutions to zero-shot or few-shot learning by grasping a similarity between vision and language features for image and text pairs. Considering that large pre-trained vision-language models may contain valuable side textual information for minor classes, we propose to leverage text supervision to tackle the challenge of long-tailed learning. Concretely, we propose a novel text-guided mixup technique that takes advantage of the semantic relations between classes recognized by the pre-trained text encoder to help alleviate the long-tailed problem. Our empirical study on benchmark long-tailed tasks demonstrates the effectiveness of our proposal with a theoretical guarantee. Our code is available at https://github.com/rsamf/text-guided-mixup.