Bridging the Sim-to-Real GAP for Underwater Image Segmentation

Luiza Ribeiro Marnet, Stella Graßhof, Yuri Brodskiy, Andrzej Wasowski

Research output: Journal Article or Conference Article in JournalConference articleResearchpeer-review

Abstract

Labeling images for every new task or data pattern a model needs to learn is a significant time bottleneck in real-world applications. Moreover, acquiring the necessary data for training the models can be challenging. Ideally, one would train the models with simulated images and adapt them for the desired real tasks using the least possible amount of data. Active learning can be used to solve this problem with minimal effort. In this work, we train SegFormer for pipeline segmentation with synthetic images from an underwater simulated environment and fine-tune the model with real underwater pipeline images recorded in a marina. The evaluation shows that selecting real data with active learning for fine-tuning the model gives better results than randomly selecting the images. As part of the work, we release the dataset recorded in the marina, MarinaPipe, which will be publicly available.
Original languageEnglish
JournalOceans Conference
DOIs
Publication statusPublished - May 2024
EventOCEANS 2024 - Singapore, Singapore
Duration: 15 Apr 202418 Apr 2024
https://singapore24.oceansconference.org/

Conference

ConferenceOCEANS 2024
Country/TerritorySingapore
CitySingapore
Period15/04/202418/04/2024
Internet address

Keywords

  • active learning
  • computer vision
  • Underwater inspection
  • image segmentation
  • sim-to-real

Fingerprint

Dive into the research topics of 'Bridging the Sim-to-Real GAP for Underwater Image Segmentation'. Together they form a unique fingerprint.

Cite this