Fine-grained Categorization and Dataset Bootstrapping using Deep Metric Learning with Humans in the Loop

TL;DR

This paper introduces an iterative deep metric learning framework with human-in-the-loop for fine-grained categorization, effectively expanding datasets and improving classification accuracy on challenging visual categories.

Contribution

The work presents a novel human-in-the-loop dataset bootstrapping approach using deep metric learning to handle data scarcity and high intra-class variance in fine-grained categorization.

Findings

Achieved significant performance improvements through dataset expansion.

Demonstrated state-of-the-art results on fine-grained datasets.

Effectively handled high intra-class variance with learned feature embeddings.

Abstract

Existing fine-grained visual categorization methods often suffer from three challenges: lack of training data, large number of fine-grained categories, and high intraclass vs. low inter-class variance. In this work we propose a generic iterative framework for fine-grained categorization and dataset bootstrapping that handles these three challenges. Using deep metric learning with humans in the loop, we learn a low dimensional feature embedding with anchor points on manifolds for each category. These anchor points capture intra-class variances and remain discriminative between classes. In each round, images with high confidence scores from our model are sent to humans for labeling. By comparing with exemplar images, labelers mark each candidate image as either a "true positive" or a "false positive". True positives are added into our current dataset and false positives are regarded as "hard negatives" for our metric learning model. Then the model is retrained with an expanded dataset and hard negatives for the next round. To demonstrate the effectiveness of the proposed framework, we bootstrap a fine-grained flower dataset with 620 categories from Instagram images. The proposed deep metric learning scheme is evaluated on both our dataset and the CUB-200-2001 Birds dataset. Experimental evaluations show significant performance gain using dataset bootstrapping and demonstrate state-of-the-art results achieved by the proposed deep metric learning methods.