Point Label Aware Superpixels for Multi-species Segmentation of Underwater Imagery

TL;DR

This paper introduces a novel point label aware superpixel method for propagating sparse labels in underwater coral imagery, significantly improving segmentation accuracy and efficiency for ecological monitoring.

Contribution

It presents a new label propagation technique leveraging sparse point labels and learned features, enhancing coral reef segmentation accuracy and reducing computational costs.

Findings

Outperforms prior methods with 3.62% higher pixel accuracy

Achieves 8.35% better mean IoU in label propagation

Reduces computation time by 76%

Abstract

Monitoring coral reefs using underwater vehicles increases the range of marine surveys and availability of historical ecological data by collecting significant quantities of images. Analysis of this imagery can be automated using a model trained to perform semantic segmentation, however it is too costly and time-consuming to densely label images for training supervised models. In this letter, we leverage photo-quadrat imagery labeled by ecologists with sparse point labels. We propose a point label aware method for propagating labels within superpixel regions to obtain augmented ground truth for training a semantic segmentation model. Our point label aware superpixel method utilizes the sparse point labels, and clusters pixels using learned features to accurately generate single-species segments in cluttered, complex coral images. Our method outperforms prior methods on the UCSD Mosaics dataset by 3.62% for pixel accuracy and 8.35% for mean IoU for the label propagation task, while reducing computation time reported by previous approaches by 76%. We train a DeepLabv3+ architecture and outperform state-of-the-art for semantic segmentation by 2.91% for pixel accuracy and 9.65% for mean IoU on the UCSD Mosaics dataset and by 4.19% for pixel accuracy and 14.32% mean IoU for the Eilat dataset.