pith. sign in

arxiv: 2509.01878 · v2 · submitted 2025-09-02 · 💻 cs.RO · cs.CV· cs.LG

AI-Driven Marine Robotics: Emerging Trends in Underwater Perception and Ecosystem Monitoring

Scarlett Raine , Tobias Fischer This is my paper

Pith reviewed 2026-05-18 20:27 UTC · model grok-4.3

classification 💻 cs.RO cs.CVcs.LG
keywords marine roboticsunderwater perceptionecosystem monitoringweakly supervised learningopen-set recognitionrobust perception3D reconstructionfoundation models
0
0 comments X

The pith

Underwater challenges are driving advances in weakly supervised learning and open-set recognition that extend to general computer vision.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper argues that marine ecosystems under climate pressure need scalable AI monitoring, which has been made possible by three drivers: the necessity of ecosystem-scale observation, citizen science platforms supplying underwater datasets, and researchers shifting from saturated terrestrial vision problems. These drivers meet distinctive marine difficulties such as turbidity, cryptic species, annotation shortages, and cross-ecosystem transfer, which in turn spur progress in methods that require fewer labels, detect novel categories, and operate in poor visibility. If the account holds, underwater work is not merely borrowing existing AI but generating techniques with wider value for any domain facing data scarcity or degraded sensing conditions.

Core claim

The paper claims that three convergent drivers—environmental necessity for ecosystem-scale monitoring, democratization of underwater datasets through citizen science, and researcher migration from terrestrial computer vision—have transformed marine perception from niche application to catalyst for AI innovation, with challenges like turbidity and cryptic species detection directly advancing weakly supervised learning, open-set recognition, and robust perception under degraded conditions while enabling a shift toward targeted intervention.

What carries the argument

The three convergent drivers combined with marine-specific challenges as the mechanism that forces and channels advances in weakly supervised learning, open-set recognition, and robust perception under degraded conditions.

If this is right

  • Methods developed for low-visibility underwater scenes can be transferred to improve perception in terrestrial settings with fog, dust, or smoke.
  • Open-set recognition techniques tuned on cryptic marine species may help detect previously unseen objects or anomalies in medical or security imagery.
  • Self-supervised and foundation-model adaptations tested on marine data can lower the cost of deploying perception systems where expert labels are scarce.
  • The move toward active, targeted intervention robots in oceans suggests similar closed-loop systems for precision agriculture or disaster response on land.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If citizen-science data collection scales further, similar bottom-up dataset growth could bootstrap AI research in other label-poor scientific domains such as astronomy or ecology.
  • The pattern of extreme-environment constraints yielding general-purpose techniques may repeat in emerging areas like polar or deep-space robotics once datasets become available.
  • Hybrid systems that combine public participation with advanced perception could become standard for large-scale environmental monitoring programs worldwide.

Load-bearing premise

The premise that the three listed drivers are the primary factors that turned underwater perception into a source of new AI methods rather than a simple application domain.

What would settle it

A historical analysis or citation study showing that key papers on weakly supervised learning and open-set recognition predate or show little influence from underwater datasets or marine challenges would falsify the claimed causal link.

Figures

Figures reproduced from arXiv: 2509.01878 by Scarlett Raine, Tobias Fischer.

Figure 1
Figure 1. Figure 1: Visual outline. We discuss the three main drivers [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 3
Figure 3. Figure 3: Cryptic marine species. Left: a monkfish, center: [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
read the original abstract

Marine ecosystems face increasing pressure due to climate change, driving the need for scalable, AI-powered monitoring solutions to inform effective conservation and restoration efforts. This paper examines the rapid emergence of underwater AI as a major research frontier and analyzes the factors that have transformed marine perception from a niche application into a catalyst for AI innovation. We identify three convergent drivers: i) environmental necessity for ecosystem-scale monitoring, ii) democratization of underwater datasets through citizen science platforms, and iii) researcher migration from saturated terrestrial computer vision domains. Our analysis reveals how unique underwater challenges - turbidity, cryptic species detection, expert annotation bottlenecks, and cross-ecosystem generalization - are driving fundamental advances in weakly supervised learning, open-set recognition, and robust perception under degraded conditions. We survey emerging trends in datasets, scene understanding and 3D reconstruction, highlighting the paradigm shift from passive observation toward AI-driven, targeted intervention capabilities. The paper demonstrates how underwater constraints are pushing the boundaries of foundation models, self-supervised learning, and perception, with methodological innovations that extend far beyond marine applications to benefit general computer vision, robotics, and environmental monitoring.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The manuscript is a survey paper examining the emergence of AI-driven marine robotics, with a focus on underwater perception and ecosystem monitoring. It identifies three convergent drivers—environmental necessity for ecosystem-scale monitoring, democratization of underwater datasets via citizen science, and researcher migration from terrestrial computer vision—as having transformed marine perception into a catalyst for AI advances. The paper analyzes how underwater challenges (turbidity, cryptic species, annotation bottlenecks, cross-ecosystem generalization) drive progress in weakly supervised learning, open-set recognition, and robust perception, while surveying trends in datasets, scene understanding, 3D reconstruction, and a shift toward targeted intervention capabilities with broader implications for general computer vision and robotics.

Significance. If the central analysis holds, the survey offers a timely synthesis that positions marine robotics challenges as a source of methodological innovation in AI, with potential benefits extending to foundation models, self-supervised learning, and environmental monitoring. This framing could help researchers in robotics and computer vision recognize cross-domain opportunities amid climate-driven needs for scalable ecosystem monitoring.

major comments (1)
  1. [Abstract] Abstract: The claim that the three convergent drivers are the primary factors transforming marine perception and driving advances in weakly supervised learning, open-set recognition, and robust perception under degraded conditions is presented as an analytical result. However, this rests on narrative synthesis of the literature without quantitative support such as publication growth curves correlated to specific driver events, dataset provenance statistics, or before/after comparisons of method adoption. This attribution is load-bearing for the paper's strongest claim but remains an organizing hypothesis rather than demonstrated causality.
minor comments (2)
  1. [Abstract] The abstract is information-dense; splitting the description of challenges and resulting advances into separate sentences would improve readability without altering content.
  2. [Survey sections] Ensure that all trend descriptions in the survey sections are accompanied by explicit references to the underlying literature sources to allow readers to trace the cited developments.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their thoughtful and constructive review, which recognizes the timeliness of the survey. We have addressed the major comment by revising the abstract and introduction to clarify that the three drivers represent trends identified through literature synthesis rather than demonstrated causal relationships. These changes preserve the paper's core framing while improving precision.

read point-by-point responses

  1. Referee: The claim that the three convergent drivers are the primary factors transforming marine perception and driving advances in weakly supervised learning, open-set recognition, and robust perception under degraded conditions is presented as an analytical result. However, this rests on narrative synthesis of the literature without quantitative support such as publication growth curves correlated to specific driver events, dataset provenance statistics, or before/after comparisons of method adoption. This attribution is load-bearing for the paper's strongest claim but remains an organizing hypothesis rather than demonstrated causality.

    Authors: We appreciate this observation and agree that the identification of the three drivers is based on qualitative synthesis of trends in the literature rather than quantitative causal analysis. In the revised manuscript, we have updated the abstract to describe these as 'key convergent trends identified in the literature' and revised the corresponding analysis sections to emphasize interpretive synthesis. We have added a brief discussion of supporting bibliometric indicators, such as growth in underwater robotics publications and citizen-science dataset releases, with appropriate citations. We now explicitly state that establishing strict causality lies beyond the scope of a survey and frame the drivers as hypothesized contributors based on observed co-occurrence with methodological advances. revision: yes

Circularity Check

0 steps flagged

Survey organizes external literature around three drivers with no self-referential reductions or fitted predictions

full rationale

This is a literature survey paper whose central claims consist of identifying three convergent drivers and linking underwater challenges to advances in weakly supervised learning and related methods. No mathematical derivations, equations, or parameter-fitting steps are present in the abstract or described structure. The analysis draws on external citations rather than reducing any prediction or uniqueness claim to quantities defined within the paper itself or to self-citations that bear the full load of the argument. The derivation chain is therefore self-contained against external benchmarks and does not exhibit any of the enumerated circularity patterns.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The paper is a literature survey that rests on domain assumptions about environmental pressures and AI research dynamics without introducing new fitted parameters or postulated entities.

axioms (1)
  • domain assumption Marine ecosystems face increasing pressure due to climate change, driving the need for scalable monitoring.
    Opening premise of the abstract that frames the entire analysis.

pith-pipeline@v0.9.0 · 5720 in / 1316 out tokens · 39762 ms · 2026-05-18T20:27:18.273780+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

77 extracted references · 77 canonical work pages · 1 internal anchor

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

    work page

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

    work page

  3. [3]

    Alonso, I.; Yuval, M.; Eyal, G.; Treibitz, T.; and Murillo, A. C. 2019. CoralSeg: Learning coral segmentation from sparse annotations. Journal of Field Robotics, 36(8): 1456--1477

    work page 2019

  4. [4]

    R.; Helmstedt, K

    Anthony, K. R.; Helmstedt, K. J.; Bay, L. K.; Fidelman, P.; Hussey, K. E.; Lundgren, P.; Mead, D.; McLeod, I. M.; Mumby, P. J.; Newlands, M.; et al. 2020. Interventions to help coral reefs under global change—A complex decision challenge. PLOS One , 15(8): e0236399

    work page 2020

  5. [5]

    Australian Institute of Marine Science (AIMS) . 2024. ReefCloud. https://doi.org/10.25845/g5gk-ty57. Accessed: 2025-08-27

    work page doi:10.25845/g5gk-ty57 2024

  6. [6]

    J.; Kline, D

    Beijbom, O.; Edmunds, P. J.; Kline, D. I.; Mitchell, B. G.; and Kriegman, D. 2012. Automated annotation of coral reef survey images. In IEEE Conference on Computer Vision and Pattern Recognition, 1170--1177

    work page 2012

  7. [7]

    CG Labs . 2025. ReefOS: The most advanced AI for the reef. https://cglabs.coralgardeners.org/. Accessed: August 2025

    work page 2025

  8. [8]

    Chen, W.; Lei, Y.; Luo, S.; Zhou, Z.; Li, M.; and Pun, C.-M. 2024. UWFormer: U nderwater image enhancement via a semi-supervised multi-scale transformer. In IEEE International Joint Conference on Neural Networks, 1--8

    work page 2024

  9. [9]

    Citizens of the Reef . 2025. Great Reef Census: Join a new era of reef conservation. https://greatreefcensus.org/. Accessed: August 2025

    work page 2025

  10. [10]

    Connolly, R.; Herrera, C.; Rasmussen, J.; Buelow, C.; Sievers, M.; Jinks, K.; Brown, C.; Lopez-Marcano, S.; Sherman, C.; Mart \' nez-Baena, F.; et al. 2024. Estimating enhanced fish production on restored shellfish reefs using automated data collection from underwater videos. Journal of Applied Ecology, 61(4): 633--646

    work page 2024

  11. [11]

    Connolly, R.; Jinks, K.; Shand, A.; Taylor, M.; Gaston, T.; Becker, A.; and Jinks, E. 2023. Out of the shadows: Automatic fish detection from acoustic cameras. Aquatic Ecology, 57(4): 833--844

    work page 2023

  12. [12]

    M.; Jinks, K

    Connolly, R. M.; Jinks, K. I.; Herrera, C.; and Lopez-Marcano, S. 2022. Fish surveys on the move: Adapting automated fish detection and classification frameworks for videos on a remotely operated vehicle in shallow marine waters. Frontiers in Marine Science, 9: 918504

    work page 2022

  13. [13]

    L.; Carpentier, L.; Gogendeau, P.; et al

    Contini, M.; Illien, V.; Julien, M.; Ravitchandirane, M.; Russias, V.; Lazennec, A.; Chevrier, T.; Rintz, C. L.; Carpentier, L.; Gogendeau, P.; et al. 2025. Seatizen Atlas: A collaborative dataset of underwater and aerial marine imagery. Scientific Data, 12(1): 67

    work page 2025

  14. [14]

    Data Mermaid . 2025. The Data Aid your reef conservation deserves. https://datamermaid.org/. Accessed: August 2025

    work page 2025

  15. [15]

    R.; and Hughes, T

    Dietzel, A.; Bode, M.; Connolly, S. R.; and Hughes, T. P. 2021. The population sizes and global extinction risk of reef-building coral species at biogeographic scales. Nature Ecology & Evolution, 5(5): 663--669

    work page 2021

  16. [16]

    M.; Buelow, C

    Ditria, E. M.; Buelow, C. A.; Gonzalez-Rivero, M.; and Connolly, R. M. 2022. Artificial intelligence and automated monitoring for assisting conservation of marine ecosystems: A perspective. Frontiers in Marine Science, 9: 918104

    work page 2022

  17. [17]

    FathomVerse. 2025. FathomVerse. https://www.fathomverse.game/. Accessed: August 2025

    work page 2025

  18. [18]

    E.; Ganase, A.; Gonzalez-Marrero, Y.; Herrera-Reveles, A.; Kennedy, E

    Gonzalez-Rivero, M.; Beijbom, O.; Rodriguez-Ramirez, A.; Bryant, D. E.; Ganase, A.; Gonzalez-Marrero, Y.; Herrera-Reveles, A.; Kennedy, E. V.; Kim, C. J.; Lopez-Marcano, S.; et al. 2020. Monitoring of coral reefs using artificial intelligence: A feasible and cost-effective approach. Remote Sensing, 12(3): 489

    work page 2020

  19. [19]

    Gonz \'a lez-Rivero, M.; Bongaerts, P.; Beijbom, O.; Pizarro, O.; Friedman, A.; Rodriguez-Ramirez, A.; Upcroft, B.; Laffoley, D.; Kline, D.; Bailhache, C.; et al. 2014. The Catlin Seaview Survey--kilometre-scale seascape assessment, and monitoring of coral reef ecosystems. Aquatic Conservation: Marine and Freshwater Ecosystems, 24(S2): 184--198

    work page 2014

  20. [20]

    V.; Neal, B

    Gonz \'a lez-Rivero, M.; Rodriguez-Ramirez, A.; Beijbom, O.; Dalton, P.; Kennedy, E. V.; Neal, B. P.; Vercelloni, J.; Bongaerts, P.; Ganase, A.; Bryant, D. E.; et al. 2019. Seaview survey photo-quadrat and image classification dataset

    work page 2019

  21. [21]

    Gorry, B.; Fischer, T.; Milford, M.; and Fontan, A. 2025. Image-based relocalization and alignment for long-term monitoring of dynamic underwater environments. In IEEE/RSJ Conference on Intelligent Robots and Systems

    work page 2025

  22. [22]

    Hanlon, R. 2007. Cephalopod dynamic camouflage. Current biology, 17(11): R400--R404

    work page 2007

  23. [23]

    J.; McGregor, H

    Henley, B. J.; McGregor, H. V.; King, A. D.; Hoegh-Guldberg, O.; Arzey, A. K.; Karoly, D. J.; Lough, J. M.; DeCarlo, T. M.; and Linsley, B. K. 2024. Highest ocean heat in four centuries places Great Barrier Reef in danger. Nature, 632(8024): 320--326

    work page 2024

  24. [24]

    Hu, S.; and Liu, Q. 2025. Fast underwater scene reconstruction using multi-view stereo and physical imaging. Neural Networks, 107568

    work page 2025

  25. [25]

    Hu, W.-C.; Chen, L.-B.; Wang, B.-H.; Li, G.-W.; and Huang, X.-R. 2023. Design and implementation of a full-time artificial intelligence of things-based water quality inspection and prediction system for intelligent aquaculture. IEEE Sensors Journal, 24(3): 3811--3821

    work page 2023

  26. [26]

    Katija, K.; Orenstein, E.; Schlining, B.; Lundsten, L.; Barnard, K.; Sainz, G.; Boulais, O.; Cromwell, M.; Butler, E.; Woodward, B.; et al. 2022. FathomNet: A global image database for enabling artificial intelligence in the ocean. Scientific Reports, 12(1): 15914

    work page 2022

  27. [27]

    King Abdullah University of Science and Technology . 2025. Adoption of AI to accelerate world's largest coral restoration project. https://www.kaust.edu.sa/en/news/adoption-of-ai-to-accelerate-world's-largest-coral-restoration-project/. Accessed: August 2025

    work page 2025

  28. [28]

    E.; and Gill, S

    Kohler, K. E.; and Gill, S. M. 2006. Coral Point Count with Excel extensions (CPCe): A Visual Basic program for the determination of coral and substrate coverage using random point count methodology. Computers & Geosciences, 32(9): 1259--1269

    work page 2006

  29. [29]

    S.; Mateo, M.- \'A .; Serrano, O.; and Rozaimi, M

    Lavery, P. S.; Mateo, M.- \'A .; Serrano, O.; and Rozaimi, M. 2013. Variability in the carbon storage of seagrass habitats and its implications for global estimates of blue carbon ecosystem service. PLOS One , 8(9): e73748

    work page 2013

  30. [30]

    L.; Chartrand, K

    Lawson, C. L.; Chartrand, K. M.; Roelfsema, C. M.; Kolluru, A.; and Mumby, P. J. 2025. Broadscale reconnaissance of coral reefs from citizen science and deep learning. Environmental Monitoring and Assessment, 197(7): 1--21

    work page 2025

  31. [31]

    A.; Figueira, W

    Lechene, M. A.; Figueira, W. F.; Murray, N. J.; Aston, E. A.; Gordon, S. E.; and Ferrari, R. 2024. Evaluating error sources to improve precision in the co-registration of underwater 3D models. Ecological Informatics, 81: 102632

    work page 2024

  32. [32]

    Li, H.; Song, W.; Xu, T.; Elsig, A.; and Kulhanek, J. 2025. W ater S platting: Fast underwater 3D scene reconstruction using Gaussian splatting. In International Conference on 3D Vision

    work page 2025

  33. [33]

    Li, Z.; Chen, Q.; and Miao, J. 2025. SPMFormer: S implified Physical Model-based transformer with cross-space loss for underwater image enhancement. Knowledge-Based Systems, 113694

    work page 2025

  34. [34]

    C.; Misiuk, B.; Xu, I.; Abdulazizov, S.; Baroi, A

    Lowe, S. C.; Misiuk, B.; Xu, I.; Abdulazizov, S.; Baroi, A. R.; Bastos, A. C.; Best, M.; Ferrini, V.; Friedman, A.; Hart, D.; et al. 2025. BenthicNet: A global compilation of seafloor images for deep learning applications. Scientific Data, 12(1): 230

    work page 2025

  35. [35]

    A.; Sievers, M.; Reeves, S

    McAneney, C. A.; Sievers, M.; Reeves, S. E.; Rasmussen, J. A.; Leaning, A.; Lopez-Marcano, S.; and Connolly, R. M. 2025. Fish use of restored shellfish reefs extends beyond the reef edge. Restoration Ecology, e70064

    work page 2025

  36. [36]

    J.; Nordlund, L

    McKenzie, L. J.; Nordlund, L. M.; Jones, B. L.; Cullen-Unsworth, L. C.; Roelfsema, C.; and Unsworth, R. K. 2020. The global distribution of seagrass meadows. Environmental Research Letters, 15(7): 074041

    work page 2020

  37. [37]

    Mou, S.; Tsai, D.; and Dunbabin, M. 2022. Reconfigurable robots for scaling reef restoration. arXiv preprint arXiv:2205.04612

    work page arXiv 2022

  38. [38]

    M.; and Jenkins, G

    Murphy, H. M.; and Jenkins, G. P. 2010. Observational methods used in marine spatial monitoring of fishes and associated habitats: A review. Marine and Freshwater Research, 61(2): 236--252

    work page 2010

  39. [39]

    NASA. 2025. Nemo-Net. https://nemonet.info/. Accessed: August 2025

    work page 2025

  40. [40]

    NOAA. 2025. NOAA Digital Photo Collection. https://www.noaa.gov/digital-collections/photo-library. Accessed: August 2025

    work page 2025

  41. [41]

    R.; Fakhrurrozi; Barker, T.; Rahman, M

    Nozdrenkov, S.; Mujiyanto; Zedta, R. R.; Fakhrurrozi; Barker, T.; Rahman, M. D.; Samusamu, A. S.; Rochman, F.; Peters, L.; Rachmawati, D. R.; Johan, D. O.; Craggs, D. J.; and Sweet, P. M. 2025. sweet-corals (Revision bd316dc)

    work page 2025

  42. [42]

    Pierce, J. 2025. Click-a-Coral. https://www.zooniverse.org/projects/jordan-pierce/click-a-coral. Accessed: August 2025

    work page 2025

  43. [43]

    H.; Guan, Y

    Qiao, J. H.; Guan, Y. T.; and Zhi, J. N. 2025. Transformer-CNN hybrid network for underwater image enhancement. The Journal of Supercomputing, 81(10): 1--17

    work page 2025

  44. [44]

    QUT. 2024. Robot designed to boost coral regeneration on the Great Barrier Reef. https://www.qut.edu.au/about/faculty-of-science/news-and-events?id=194074. Accessed: August 2025

    work page 2024

  45. [45]

    W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al

    Radford, A.; Kim, J. W.; Hallacy, C.; Ramesh, A.; Goh, G.; Agarwal, S.; Sastry, G.; Askell, A.; Mishkin, P.; Clark, J.; et al. 2021. Learning transferable visual models from natural language supervision. In International Conference on Machine Learning, 8748--8763. PmLR

    work page 2021

  46. [46]

    Raine, S.; Maire, F.; Suenderhauf, N.; and Fischer, T. 2024 a . Reducing label dependency for underwater scene understanding: A survey of datasets, techniques and applications. arXiv preprint arXiv:2411.11287

    work page arXiv 2024

  47. [47]

    Raine, S.; Marchant, R.; Kusy, B.; Maire, F.; and Fischer, T. 2022. Point label aware superpixels for multi-species segmentation of underwater imagery. IEEE Robotics and Automation Letters, 7(3): 8291--8298

    work page 2022

  48. [48]

    Raine, S.; Marchant, R.; Kusy, B.; Maire, F.; and Fischer, T. 2024 b . Image labels are all you need for coarse seagrass segmentation. In IEEE/CVF Winter Conference on Applications of Computer Vision, 5943--5952

    work page 2024

  49. [49]

    Raine, S.; Marchant, R.; Kusy, B.; Maire, F.; Sunderhauf, N.; and Fischer, T. 2024 c . Human-in-the-loop segmentation of multi-species coral imagery. In IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2723--2732

    work page 2024

  50. [50]

    Raine, S.; Marchant, R.; Moghadam, P.; Maire, F.; Kettle, B.; and Kusy, B. 2020. Multi-species seagrass detection and classification from underwater images. In IEEE Digital Image Computing: Techniques and Applications

    work page 2020

  51. [51]

    Raine, S.; Moshirian, B.; and Fischer, T. 2025. AI-driven dispensing of coral reseeding devices for broad-scale restoration of the Great Barrier Reef. arXiv preprint

    work page 2025

  52. [52]

    Raynor, J.; Orofino, S.; Costello, C.; McDonald, G.; Mayorga, J.; and Sala, E. 2025. Little-to-no industrial fishing occurs in fully and highly protected marine areas. Science, 389(6758): 392--395

    work page 2025

  53. [53]

    ReefCloud. 2025. Protecting the Future of Coral Reefs. https://reefcloud.ai/. Accessed: August 2025

    work page 2025

  54. [54]

    Remmers, T.; Boutros, N.; Wyatt, M.; Gordon, S.; Toor, M.; Roelfsema, C.; Fabricius, K.; Grech, A.; Lechene, M.; and Ferrari, R. 2025. RapidBenthos: Automated segmentation and multi-view classification of coral reef communities from photogrammetric reconstruction. Methods in Ecology and Evolution, 16(2): 427--441

    work page 2025

  55. [55]

    Sauder, J.; Banc-Prandi, G.; Meibom, A.; and Tuia, D. 2024. Scalable semantic 3D mapping of coral reefs with deep learning. Methods in Ecology and Evolution, 15(5): 916--934

    work page 2024

  56. [56]

    Sauder, J.; Domazetoski, V.; Banc-Prandi, G.; Perna, G.; Meibom, A.; and Tuia, D. 2025. The Coralscapes dataset: Semantic scene understanding in coral reefs. arXiv preprint arXiv:2503.20000

    work page arXiv 2025

  57. [57]

    DINOv3

    Sim \'e oni, O.; Vo, H. V.; Seitzer, M.; Baldassarre, F.; Oquab, M.; Jose, C.; Khalidov, V.; Szafraniec, M.; Yi, S.; Ramamonjisoa, M.; et al. 2025. DINOv3. arXiv preprint arXiv:2508.10104

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  58. [58]

    S.; and Nayyar, A

    Singh, M.; Sahoo, K. S.; and Nayyar, A. 2022. Sustainable iot solution for freshwater aquaculture management. IEEE Sensors Journal, 22(16): 16563--16572

    work page 2022

  59. [59]

    Souter, D.; Planes, S.; Wicquart, J.; Logan, M.; Obura, D.; and Staub, F., eds. 2021. Status of Coral Reefs of the World: 2020 Report. Global Coral Reef Monitoring Network (GCRMN) and International Coral Reef Initiative (ICRI). DOI: 10.59387/WOTJ9184

    work page doi:10.59387/wotj9184 2021

  60. [60]

    Squidle. 2025. IMOS-UMI Squidle+ A tool for managing, exploring & annotating images, video & large-scale mosaics. https://squidle.org/. Accessed: August 2025

    work page 2025

  61. [61]

    Tidal. 2025. AI for the next generation of sustainable aquaculture. https://www.tidalx.ai/en. Accessed: August 2025

    work page 2025

  62. [62]

    A.; Lan, H.-Y.; Cheng, S.-C.; Chang, C.-C.; Lin, S.-S.; Zhang, K.-X.; Lu, H.-Y.; Cheng, C.-Y.; and Hsieh, Y.-Z

    Ubina, N. A.; Lan, H.-Y.; Cheng, S.-C.; Chang, C.-C.; Lin, S.-S.; Zhang, K.-X.; Lu, H.-Y.; Cheng, C.-Y.; and Hsieh, Y.-Z. 2023. Digital twin-based intelligent fish farming with Artificial Intelligence Internet of Things (AIoT). Smart Agricultural Technology, 5: 100285

    work page 2023

  63. [63]

    L.; and Segal-Rozenhaimer, M

    Van den Bergh, J.; Chirayath, V.; Li, A.; Torres-P \'e rez, J. L.; and Segal-Rozenhaimer, M. 2021. NeMO-net--gamifying 3D labeling of multi-modal reference datasets to support automated marine habitat mapping. Frontiers in Marine Science, 8: 645408

    work page 2021

  64. [64]

    Vercelloni, J.; Santos-Fern \'a ndez, E.; and Mengersen, K. 2023. Virtual Reef Diver: Enabling People to Help Protect the Great Barrier Reef. Citizen Science: Theory and Practice, 8(1)

    work page 2023

  65. [65]

    Virtual Reef Diver . 2025. Virtual Reef Diver: Become a Citizen Scientist. https://www.virtualreef.org.au/. Accessed: August 2025

    work page 2025

  66. [66]

    Wang, B.; Xu, H.; Jiang, G.; Yu, M.; Chen, Y.; Ding, L.; Zhang, X.; and Luo, T. 2023. Underwater image co-enhancement based on physical-guided transformer interaction. Displays, 79: 102505

    work page 2023

  67. [67]

    Wang, H.; Anantrasirichai, N.; Zhang, F.; and Bull, D. 2025. UW-GS: Distractor-aware 3D Gaussian splatting for enhanced underwater scene reconstruction. In IEEE/CVF Winter Conference on Applications of Computer Vision, 3280--3289

    work page 2025

  68. [68]

    Wille, M.; Fischer, T.; and Raine, S. 2025. Are all marine species created equal? Performance disparities in underwater object detection. arXiv 2508.18729

    work page arXiv 2025

  69. [69]

    Woods Hole Oceanographic Institution . 2022. Digital Reefs awarded \ 5 million. https://www.whoi.edu/press-room/news-release/digital-reefs-awarded-5-million/. Accessed: August 2025

    work page 2022

  70. [70]

    Wyatt, M.; Hickey, S.; Radford, B.; Gonzalez-Rivero, M.; Boutros, N.; Callow, N.; Ryan, N.; Chennu, A.; Bennamoun, M.; and Gilmour, J. 2025. Safe AI for coral reefs: Benchmarking out-of-distribution detection algorithms for coral reef image surveys. Ecological Informatics, 103207

    work page 2025

  71. [71]

    J.; and Girdhar, Y

    Yang, D.; Leonard, J. J.; and Girdhar, Y. 2025. SeaSplat: Representing underwater scenes with 3D Gaussian splatting and a physically grounded image formation model. In IEEE International Conference on Robotics and Automation

    work page 2025

  72. [72]

    Zheng, Z.; Chen, Y.; Zeng, H.; Vu, T.-A.; Hua, B.-S.; and Yeung, S.-K. 2024 a . MarineInst: A foundation model for marine image analysis with instance visual description. European Conference on Computer Vision

    work page 2024

  73. [73]

    H.; Ang, P.; Chui, A

    Zheng, Z.; Liang, H.; Hua, B.-S.; Wong, Y. H.; Ang, P.; Chui, A. P. Y.; and Yeung, S.-K. 2024 b . CoralSCOP: Segment any coral image on this planet. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 28170--28180

    work page 2024

  74. [74]

    Zheng, Z.; Wong, K.; Binh-Son, H.; Jianbo, S.; and Sai-Kit, Y. 2025. CoralSRT: R evisiting coral reef semantic segmentation by feature rectifying via self-supervised guidance. In Proceedings of the IEEE International Conference on Computer Vision

    work page 2025

  75. [75]

    Zheng, Z.; Xie, Y.; Liang, H.; Yu, Z.; and Yeung, S.-K. 2023 a . CoralVOS: Dataset and benchmark for coral video segmentation. arXiv preprint arXiv:2310.01946

    work page arXiv 2023

  76. [76]

    Zheng, Z.; Zhang, J.; Vu, T.-A.; Diao, S.; Tim, Y. H. W.; and Yeung, S.-K. 2023 b . MarineGPT: Unlocking secrets of ocean to the public. arXiv preprint arXiv:2310.13596

    work page arXiv 2023

  77. [77]

    Zhong, J.; Li, M.; Zhang, H.; and Qin, J. 2023. Combining photogrammetric computer vision and semantic segmentation for fine-grained understanding of coral reef growth under climate change. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 186--195

    work page 2023