Empowering DINO Representations for Underwater Instance Segmentation via Aligner and Prompter

TL;DR

This paper introduces DiveSeg, a novel underwater instance segmentation framework that enhances DINO features with domain-specific style embedding and object priors, achieving state-of-the-art results on underwater datasets.

Contribution

The paper presents DiveSeg, combining AquaStyle Aligner and ObjectPrior Prompter to adapt DINO for underwater segmentation, a novel approach for domain-specific fine-tuning.

Findings

Achieves state-of-the-art performance on UIIS and USIS10K datasets.

Effectively incorporates underwater color style features into DINO.

Utilizes object-level priors for improved instance segmentation.

Abstract

Underwater instance segmentation (UIS), integrating pixel-level understanding and instance-level discrimination, is a pivotal technology in marine resource exploration and ecological protection. In recent years, large-scale pretrained visual foundation models, exemplified by DINO, have advanced rapidly and demonstrated remarkable performance on complex downstream tasks. In this paper, we demonstrate that DINO can serve as an effective feature learner for UIS, and we introduce DiveSeg, a novel framework built upon two insightful components: (1) The AquaStyle Aligner, designed to embed underwater color style features into the DINO fine-tuning process, facilitating better adaptation to the underwater domain. (2) The ObjectPrior Prompter, which incorporates binary segmentation-based prompts to deliver object-level priors, provides essential guidance for instance segmentation task that requires both object- and instance-level reasoning. We conduct thorough experiments on the popular UIIS and USIS10K datasets, and the results show that DiveSeg achieves the state-of-the-art performance. Code: https://github.com/ettof/Diveseg.