Mapping the Reef: Underwater 3D Reconstruction for Coral Ecosystems

• Feb 12, 2026 - Workshop proposal submission!

Long-term 3D mapping across repeated site visits is a key capability for automating the monitoring of coral reef ecosystems, enabling applications such as structural change detection, ecosystem health assessment, and targeted intervention. Coral reefs are a particularly compelling and challenging domain: they are among the most biodiverse and valuable marine ecosystems, yet are increasingly threatened by climate change and other stressors. Monitoring efforts commonly involve repeated surveys spanning months or years., and converting these observations into reliable, quantitative evidence requires mapping methods that remain consistent across time, conditions, and platforms.

This workshop aims to bring together researchers and practitioners to share, compare, and propose techniques that can enable scalable reef monitoring. We will explore approaches that improve the ability to build useful, repeatable maps of reefs: optical, acoustic, and hybrid reconstruction; SLAM and multi-session mapping; learned and physics-informed representations; semantic and metric mapping; uncertainty estimation and quality assurance; active perception and next-best-view planning; and system-level practices for field deployment. The emphasis is on identifying what works, what fails in the wild, and what new ideas and evaluation standards are needed to turn underwater mapping into a dependable tool for reef science and conservation.

We invite short papers of novel or recently published research relevant to the topics of the workshop.

• Short papers are 2+n pages (2 pages content + references)
• Submissions must follow the IEEE Conference double column format
• All accepted papers will be presented as posters at the workshop and published on the workshop website.
• Please indicate whether your paper falls into the ‘novel’ or ‘previously published’ category. Novel research papers are encouraged and can expect more substantial review feedback on their work. This is provided as a service to authors of novel papers and does not diminish the chance of acceptance.
• All accepted submissions will be considered for the best presentation award, where 3 finalists will be selected for 10-minute plenary presentations. While all submissions are eligible, novelty will be considered in finalist selection.
• Submissions are single blind and will be reviewed by members of the (extended) workshop committee.
• Submissions can optionally be accompanied by a video.
• All accepted submissions will be considered for three awards: Best Paper, Best Poster, and Best Application/Deployed System, each receiving a 150 USD gift card.

Invited topics

We invite submissions including, but not limited to:

• Geometric Mapping (SLAM/VO, multi-session): How do we get metric, repeatable, globally consistent long-term reef maps without GNSS under refraction/turbidity/lighting change/currents/limited loop closures—and when is pinhole enough vs refractive/ray-based?
• Learning-Based Approaches: How can learning deliver robust reconstruction + cross-session alignment under major change and underwater domain shift/data scarcity/visual degradation—and what reef semantics (morphology, live/dead, algae, rubble, bleaching/disease) should be baked in for monitoring?
• Autonomy and Active Perception: Mission planning, next-best-view strategies, reconstructability prediction, and adaptive behaviors for reef surveys.
• Underwater Scene Representations (NeRF/splatting/implicit maps): What representations best balance metric accuracy, interpretability, scalability, and long-term repeatability for reef monitoring—and how can foundation models and neural scenes (NeRFs, Gaussian splatting, implicit SDFs) be adapted to low-label underwater data to be practical, trustworthy, and usable by marine scientists?
• Multi-Modal Sensing + Fusion: Which sensor combinations (e.g., cameras + imaging sonar/multibeam/SSS, structured light, polarization, hyperspectral/fluorescence, water-quality) and fusion strategies yield reliable reef maps when any single modality fails—and how should suites be designed, calibrated, and deployed for robust underwater perception.
• Benchmarking Strategies for Underwater: Quantitative evaluation methods, uncertainty modeling, and benchmarking frameworks for underwater field experiments.
• From Robotics to Reef Science (field-ready systems): What representations and performance thresholds make reef mapping deployable, and how can robots predict reconstructability, report uncertainty, and adapt via active perception (next-best-view, re-scan, modality switching) as conditions degrade—while supporting scalable evaluation/ground-truthing without traditional reference systems?

Time	Planned Event	Speaker
08:00	Workshop kick-off and overview
08:00	Talk 1- Wildflow: Modelling natural ecosystems	Sergei Nozdrenkov
08:30	Talk 2 - MOANA: Multi-radar dataset for maritime odometry and autonomous navigation applications	Ayoung Kim
09:00	Talk 3 - Multi-Modal SLAM for Reliable Underwater Exploration	Ioannis Rekleitis
09:30	Coffee Break and Poster Session
10:00	Spotlight Talks: Best Paper, Best Poster and Best Practical Contribution
10:30	Talk 4 - From prototype to practice: real-world integration of robotics into marine management	Vincent Raoult
11:00	Talk 5 - Resilient Vision-based Underwater Autonomy	Kostas Alexis
11:30	Talk 6 - Effective scene representations for marine scientists	Maren Toor
12:00	Mapping the Reef: Underwater 3D Reconstruction for Coral Ecosystems