PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 2, 2026Frontiers in Marine Science2 citationsOpen Access

Squidle+: a collaborative platform to manage, discover and annotate marine imagery

View Full Paper
AFAriell FriedmanJMJacquomo MonkOPOscar Pizarro

Key Points

  • The aim is to present Squidle+, a platform for managing and annotating marine imagery efficiently.
  • Developed a web-based collaborative platform for marine imagery management.
  • Integrated cloud storage for direct imagery access without data duplication.
  • Created annotation workflows and tools for advanced data analysis.
  • Implemented user groups for collaboration with customizable permissions.
  • Established a framework for translating multiple annotation vocabularies.
  • Squidle+ serves as the largest open repository of georeferenced marine images with annotations.
  • Improved workflows enhance the findability and accessibility of marine image data.
  • The system supports large-scale ML training and facilitates data reuse across projects.

Abstract

The rapid growth of marine imaging has outpaced our ability to efficiently analyse the imagery, creating challenges in data management, collaboration, and standardisation. This paper presents Squidle+, a web-based, collaborative platform for the end-to-end management, delivery, discovery, and annotation of marine imagery. Squidle+ provides a centralised portal and annotation repository while linking to imagery hosted on pre-existing cloud storage, eliminating data transfer and duplication. The system features a user-friendly interface with map-based exploration tools, advanced annotation workflows, and integrated analytics through a comprehensive API back-end. Collaboration is managed through user groups with granular permissions, while integrated QA/QC tools enable cross-validation between human annotators and Machine Learning (ML) algorithms. A key innovation is a framework to translate between multiple standardised or user-defined annotation vocabularies. This gives users the flexibility to construct data sets that target specific scientific questions and facilitates data reuse, cross-project syntheses, large-scale ML training, and broad summaries that can be fed into national-level reporting. Squidle+ has been developed in close collaboration with an active user community and currently contains datasets from several platforms and operators around the world. It is currently the largest known repository of openly accessible georeferenced marine images with associated annotations. Squidle+ streamlines complex workflows and significantly enhances the Findability, Accessibility, Interoperability, and Reusability (FAIR) of marine image data.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Friedman et al. (2026) studied this question.

synapsesocial.com/papers/6980fe35c1c9540dea8101a8https://doi.org/10.3389/fmars.2025.1677103
Ask AI
Helpful
Bookmark
Share
View Full Paper