Towards Scaling Marine Perception with Synthetic Data: OceanSim SDG Pipeline

Forum topic · 小凯 · 2026-09-20

Summary

A new paper (arXiv:2609.20680) from researchers including Haoyu Ma and Katherine A. Skinner extends OceanSim, an IsaacSim-based underwater perception simulator, with a Synthetic Data Generation (SDG) pipeline. The pipeline addresses the scarcity of labeled real-world underwater training data, which is expensive and laborious to collect. It enables users to generate large-scale, automatically labeled, photorealistic datasets with configurable scene appearance, structure, and sensor settings. The authors evaluate the pipeline on a real-world sea urchin detection task, studying how different forms of synthetic scene variation affect sim-to-real performance. Based on the experiments, the paper discusses key insights, main limitations of the current pipeline, and future directions for improving underwater rendering fidelity, scene diversity, and sim-to-real generalization evaluation. Open-source code is available at https://github.com/umfieldrobotics/OceanSim. This work offers a practical approach for scaling machine learning in challenging underwater environments.

Overview

Research area: Computer Vision (CV) Authors: Haoyu Ma, Onur Bagoren, Anja Sheppard, Elias Fandi, Ashrith Edukulla, Tanner Aslan, Natasha Sieh, Jingyu Song, Katherine A. Skinner Published: 2026-09-17 arXiv: 2609.20680

Abstract

Scalable machine learning in challenging underwater environments is strongly limited by the lack of labeled real-world training data. This data is often expensive and laborious to gather, making large-scale real-world data challenging to gather and curate. However, simulated data can help close the gap, enabling many learning-based tasks for underwater perception.

In this work, the authors extend OceanSim, an IsaacSim-based underwater perception simulator, with a Synthetic Data Generation (SDG) pipeline for training models to be used in underwater scenarios. The proposed pipeline enables users to generate large, automatically labeled, photorealistic datasets with configurable scene appearance, structure, and sensor settings.

The pipeline is evaluated on a real-world sea urchin detection task, studying how different forms of synthetic scene variation impact sim-to-real performance. Based on the experiments, the paper discusses insights from the results, the main limitations of the current pipeline, and future directions to improve underwater rendering fidelity, scene diversity, and sim-to-real generalization evaluation.

Resources

Paper: https://arxiv.org/abs/2609.20680
Code: https://github.com/umfieldrobotics/OceanSim