Frozen Feature Augmentation for Few-Shot Image Classification

2024/03/15 by Andreas Bär, Bär, Andreas, Neil Houlsby +5 · 5 citations
Computer Science · Engineering · #Advanced Image Processing Techniques #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Image Processing Techniques and Applications

paper · pdf · doi:10.48550/arxiv.2403.10519

openalex publication_date 2024/03/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Training a linear classifier or lightweight model on top of pretrained vision model outputs, so-called 'frozen features', leads to impressive performance on a number of downstream few-shot tasks. Currently, frozen features are not modified during training. On the other hand, when networks are trained directly on images, data augmentation is a standard recipe that improves performance with no substantial overhead. In this paper, we conduct an extensive pilot study on few-shot image classification that explores applying data augmentations in the frozen feature space, dubbed 'frozen feature augmentation (FroFA)', covering twenty augmentations in total. Our study demonstrates that adopting a deceptively simple pointwise FroFA, such as brightness, can improve few-shot performance consistently across three network architectures, three large pretraining datasets, and eight transfer datasets.

Cited by

• A multimodal whole-slide foundation model for pathology
• Denoised Diffusion for Object-Focused Image Augmentation
• Slot Attention-based Feature Filtering for Few-Shot Learning
• Color as the Impetus: Transforming Few-Shot Learner
• Group Relative Augmentation for Data Efficient Action Detection

Related

• Pushing the Limits of Simple Pipelines for Few-Shot Learning: External Data and Fine-Tuning Make a Difference
• Few-shot Image Classification: Just Use a Library of Pre-trained Feature Extractors and a Simple Classifier
• An empirical study of pretrained representations for few-shot classification
• Unsupervised Embedding Adaptation via Early-Stage Feature Reconstruction for Few-Shot Classification
• Dense Classification and Implanting for Few-Shot Learning