Mask-Guided Feature Extraction and Augmentation for Ultra-Fine-Grained Visual Categorization

TL;DR

This paper introduces a mask-guided feature extraction and augmentation technique for ultra-fine-grained visual categorization, addressing challenges of small sample sizes and minimal inter-class variance to improve discriminative feature learning.

Contribution

The proposed method effectively extracts and augments discriminative regions using minimal training data, enhancing ultra-fine-grained classification performance.

Findings

Consistently outperforms ten benchmark methods

Improves feature discriminability visually and quantitatively

Requires only small target region samples for training

Abstract

While the fine-grained visual categorization (FGVC) problems have been greatly developed in the past years, the Ultra-fine-grained visual categorization (Ultra-FGVC) problems have been understudied. FGVC aims at classifying objects from the same species (very similar categories), while the Ultra-FGVC targets at more challenging problems of classifying images at an ultra-fine granularity where even human experts may fail to identify the visual difference. The challenges for Ultra-FGVC mainly comes from two aspects: one is that the Ultra-FGVC often arises overfitting problems due to the lack of training samples; and another lies in that the inter-class variance among images is much smaller than normal FGVC tasks, which makes it difficult to learn discriminative features for each class. To solve these challenges, a mask-guided feature extraction and feature augmentation method is proposed in this paper to extract discriminative and informative regions of images which are then used to augment the original feature map. The advantage of the proposed method is that the feature detection and extraction model only requires a small amount of target region samples with bounding boxes for training, then it can automatically locate the target area for a large number of images in the dataset at a high detection accuracy. Experimental results on two public datasets and ten state-of-the-art benchmark methods consistently demonstrate the effectiveness of the proposed method both visually and quantitatively.