Domain Adaptation from Generated Multi-Weather Images for Unsupervised Maritime Object Classification

TL;DR

This paper introduces a domain adaptation method utilizing generated multi-weather maritime images and vision-language models to improve unsupervised maritime object classification, especially for rare categories and weather conditions.

Contribution

It presents a new dataset AIMO created with generative models, and a domain adaptation approach that leverages CLIP and curriculum learning to enhance classification in real-world maritime images.

Findings

Significant accuracy improvements for rare object categories.

Effective handling of diverse weather conditions.

Enhanced generalization across domain shifts.

Abstract

The classification and recognition of maritime objects are crucial for enhancing maritime safety, monitoring, and intelligent sea environment prediction. However, existing unsupervised methods for maritime object classification often struggle with the long-tail data distributions in both object categories and weather conditions. In this paper, we construct a dataset named AIMO produced by large-scale generative models with diverse weather conditions and balanced object categories, and collect a dataset named RMO with real-world images where long-tail issue exists. We propose a novel domain adaptation approach that leverages AIMO (source domain) to address the problem of limited labeled data, unbalanced distribution and domain shift in RMO (target domain), enhance the generalization of source features with the Vision-Language Models such as CLIP, and propose a difficulty score for curriculum learning to optimize training process. Experimental results shows that the proposed method significantly improves the classification accuracy, particularly for samples within rare object categories and weather conditions. Datasets and codes will be publicly available at https://github.com/honoria0204/AIMO.