Pith. sign in

REVIEW 4 major objections 6 minor 39 references

ZooplanktonBench: A Geo-Aware Zooplankton Recognition and Classification Dataset from Marine Observations

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read ZooplanktonBench offers 285,733 expert-labeled zooplankton images with depth and location metadata, and shows current detectors still stumble on separating living animals from marine snow.

desk verdict A useful dataset undermined by a contaminated negative class in its central task; the binary mAP is not living-vs-marine-snow. read the letter →

arxiv 2505.18477 v3 pith:6I35572T submitted 2025-05-24 cs.CV

Fukun Liu , Adam T. Greer , Gengchen Mai , Jin Sun This is my paper
classification cs.CV
keywords ComputervisionObjectdetectionFine-grainedclassificationVideoanalysisMarinesciencezooplanktonbenchmarksnowgeospatialmetadata
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

ZooplanktonBench is a benchmark dataset built from 374,378 shadowgraph images taken by a towed camera in the northern Gulf of Mexico at three depths, from which experts cropped and identified 285,733 zooplankton in 17 broad categories. The paper's central claim is that this resource, together with its depth and oceanographic metadata, defines a set of computer vision tasks — fine-grained detection and classification, living-zooplankton versus marine-snow detection, and tracking — that are realistic for marine monitoring and challenging for current models. Baseline results back that claim: a fine-tuned YOLOv8 detector reaches mAP50 of 0.434 on the mixed-depth fine-grained task and 0.871 on the binary living-versus-marine-snow task, while a zero-shot open-set detector and GPT-4V perform far worse, and adding geo-metadata to GPT-4V often increases errors. If correct, the dataset gives marine scientists a common testbed for automated zooplankton monitoring and gives computer vision a benchmark where background clutter, small objects, and class imbalance are the obstacles. The paper also states that the benchtop video data are not annotated, so the image-plus-video track remains a proposed pipeline rather than a fully labeled training resource.

What carries the argument

The load-bearing object is the dataset itself, produced by shadowgraph imaging: the ISIIS towed vehicle photographs zooplankton as dark silhouettes against a lit background, and marine snow creates visually similar clutter. The annotation pipeline (ImageJ-assisted expert cropping into 17 classes) turns raw video frames into 285,733 labeled instances; the task definitions and the depth-based and image-plus-video tracks turn those instances into measurable problems, scored with mAP50 and mAP50-95. The image-plus-video track adds ByteTrack as a mechanism for using track consistency across five frames to distinguish living organisms from drifting marine snow.

What would settle it

Have a marine scientist fully label every living organism, including copepods and appendicularians, in a random sample of the training images, then retrain and evaluate the binary living-versus-marine-snow detector on a fully labeled test subset. If mAP50 falls well below the reported 0.871, the unlabeled living organisms in the negative class were inflating the apparent detection performance; if it stays near 0.871, the contamination is not decisive. A cheaper check is to count the detector's false positives that are identifiable copepods or appendicularians.

Watch

Extended reading notes

Core claim

The paper constructs and publicly describes ZooplanktonBench, a dataset of in situ shadowgraph imagery collected with the ISIIS towed vehicle in the northern Gulf of Mexico on 24 July 2011, at 10 m, 25 m, and 35 m depth, with geospatial and oceanographic sensor data attached (coordinates, depth, temperature, salinity, dissolved oxygen, chlorophyll fluorescence). A total of 374,378 raw images were analyzed in roughly 800 person-hours, producing 285,733 cropped organisms classified by experts into 17 categories, with fish larvae further classified to family. The proposed benchmark tasks are fine-grained detection and classification of these zooplankton and a binary living-zooplankton versus marine-snow detection task; the tracks evaluate models under cross-depth transfer and under an image-only versus image-plus-unlabeled-video training regime. Quantitative baselines on the authors' protocol show a fine-tuned YOLOv8 reaching mAP50 0.434 on the mixed-depth fine-grained task and 0.871 on the binary task, a Grounding DINO open-set detector with much lower, threshold-sensitive performance, and GPT-4V classifying only 758 of 2,188 representative crops correctly, with geo-metadata raising error messages from 84 to 323. The paper's conclusion is that the dataset captures a challenging regime—variable clutter, similar shapes, small objects, and extreme class imbalance—where general-purpose computer vision systems still fall short.

Load-bearing premise

The load-bearing premise is that every region not labeled as a zooplankton can be treated as marine snow; the paper itself notes that small living copepods and appendicularians were present in almost every image but were deliberately left unlabeled, so the binary task's negative examples are not guaranteed to be non-living.

Editorial extensions

If this is right

  • Marine scientists get a common, labeled testbed for detecting and classifying zooplankton in a productive coastal ecosystem, with depth and sensor metadata attached to every image.
  • The depth track makes ocean depth a concrete domain-shift axis: training on a mix of depths outperforms training on any single depth in the reported fine-grained results, so future geo-aware models can be scored by cross-depth transfer.
  • The image-plus-video track offers a path to use unlabeled benchtop videos, through tracking-based consistency, to reduce marine-snow false positives without expensive manual video annotation.
  • The baseline numbers set a public bar: binary living-versus-marine-snow detection is tractable at mAP50 0.871, while fine-grained classification at mAP50 0.434 and zero-shot open-set detection are far from solved.
  • The reported increase in GPT-4V errors when geo-metadata is appended means the benchmark can test whether and how environmental context should be integrated into vision-language models.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the authors state that copepods and appendicularians were present in almost every image but were not labeled, the negative class in the binary task is contaminated with living zooplankton; the reported 0.871 mAP50 is therefore likely an optimistic estimate of living-versus-non-living discrimination, and a fully labeled negative set could lower it.
  • The depth-split results imply a clean experiment the paper did not run: a model that genuinely exploits depth or location metadata should outperform a depth-agnostic baseline on held-out depths, so ZooplanktonBench's geospatial fields could anchor a benchmark for geo-aware representation learning.
  • The unannotated benchtop videos are a natural self-supervised pretraining source; measuring whether video pretraining improves the fine-grained image tasks would directly test the paper's stated motivation for including them.
  • The GPT-4V outcome suggests that appending environmental variables to a text prompt is not the right interface for geo-awareness; learned fusion of image features with sensor readings is a plausible alternative that the dataset could evaluate.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper introduces ZooplanktonBench, a benchmark dataset built from 374,378 in-situ ISIIS shadowgraph images collected in the northern Gulf of Mexico at three depths (10, 25, and 35 meters), with 285,733 cropped and human-annotated zooplankton instances assigned to 17 broad categories and accompanied by geospatial and environmental metadata. The authors define two core tasks: fine-grained detection and classification of zooplankton categories, and binary detection of living zooplankton versus marine snow. They also define tracks for cross-depth generalization and for augmenting image training with unlabeled benchtop video, and they report baselines using fine-tuned YOLOv8, zero-shot Grounding DINO, and GPT-4V with and without geo-metadata. The stated goal is to provide a challenging testbed for computer vision systems in cluttered marine environments and to support marine science applications.

Significance. If the labeling issues identified below are resolved, ZooplanktonBench has clear value as a large, real-world benchmark: it contains over a quarter-million expert annotations, includes rich geospatial and environmental context, and defines tasks that are well-motivated by marine science. The paper provides standard mAP evaluations, multiple baseline methods, and a public project website, which are useful starting points for the community. The central claim of a living-versus-marine-snow benchmark, however, is currently weakened by the fact that many living zooplankton in the images are unlabeled and are consequently treated as negative examples in the binary task. The fine-grained detection numbers are also affected by this label incompleteness. These issues are fixable by re-scoping the tasks or by correcting the ground truth, so the underlying dataset contribution remains promising.

major comments (4)
  1. [Section 3 and Section 4.1.2] The binary living-zooplankton-versus-marine-snow task is defined on contaminated labels. Section 3 states that copepods and appendicularians were present in almost every image but were not identified, while Section 4.1.2 defines the task by merging the fine-grained categories into "living zooplankton" and treating "all other distractors in the images" as marine snow. Unlabeled living copepods and appendicularians are therefore scored as marine snow (negative examples), and a detector that correctly localizes such an unlabeled animal is penalized as a false positive. Consequently, the mAP50 of 0.871 in Table 3 is not a measure of distinguishing living from non-living objects; it measures detection of the selected large taxa against an ill-defined background. The task must either be re-scoped explicitly to "large labeled zooplankton versus unlabeled background," or the ground truth must be corrected to include all living organisms, or a partial-label-aware evaluation protocol must be adopted.
  2. [Section 4.1.1 and Table 2] The fine-grained detection metrics are also affected by incomplete labeling. Because unlabeled instances of the 17 categories, notably copepods, are present in test images, any correct detection of such an instance is counted as a false positive in the mAP computation behind Table 2. This means the reported fine-grained mAP values are not an unbiased estimate of detector performance on a fully annotated benchmark. At a minimum, the authors should quantify the fraction of unlabeled category instances in the test sets or evaluate with a protocol that ignores detections matching unlabeled instances of the same category, and they should discuss how this changes the reported numbers.
  3. [Section 5.3 and Table 6] The GPT-4V evaluation lacks the reproducibility details needed for a benchmark paper. The text says "We selected a dataset of 2,188 representative zooplankton images" without specifying the selection procedure, the split between training and test (if any), or whether the selection was random, stratified, or exhaustive for a subset of classes. The "Instances" column in Table 6 sums to 2,188, which suggests the selected set may simply be all images from the listed classes, but this should be stated explicitly. The geo-metadata prompt also appears to contain a fixed set of values (latitude 29.5063167, longitude 88.5215167, depth 34.5 meters) with a note that the underlined part will be replaced; the paper must clarify whether each image receives its own corresponding metadata or whether a single metadata string is used for all images, and it must provide the exact prompt templates and selection code as supplementary material.
  4. [Section 5.1 and Table 3] The claimed benefit of the image-plus-video track is not supported by the reported results. The text states that an object assigned a track "id" is likely living zooplankton and that "we can enhance the detection accuracy by disregarding labeled objects that are untracked," but Table 3 reports Images+video mAP50 of 0.840, which is lower than the Image Only mAP50 of 0.871. The paper does not report the result of the proposed filtering step separately, nor does it give the number of tracked objects, track lengths, or the fraction of detections removed. Without an ablation, the reader cannot tell whether tracking helps, hurts, or is irrelevant; the current presentation is internally inconsistent with the stated rationale.
minor comments (6)
  1. [Abstract and Section 1] There is a missing space in the sentence "Our dataset presents unique challenges and opportunities" in the abstract and the introduction; it should read "Our dataset presents unique challenges."
  2. [Section 4.2.2] The phrase "otherwise purely relaying on images" should be "otherwise purely relying on images."
  3. [Table 1 and GPT-4V prompt] The class name "Thaliacea" in Table 1 appears as "thaliac" in the GPT-4V prompt; these should be normalized to avoid ambiguity in the reported results.
  4. [Section 5.3] The geo-metadata prompt says the image was captured "situated 34.5 meters beneath the seafloor," which is likely a typo for "below the sea surface" or "at 34.5 meters depth"; this should be corrected and the exact meaning clarified.
  5. [Section 6] The paper states that code and settings are on the project website but does not specify where the dataset itself can be downloaded, what license applies, or how to access the annotations and raw videos; this information should be included for a benchmark paper.
  6. [Table 6] The superscripts on "Correct" and "Classification" in Table 6 are not explained in the caption; the caption should define what Image and Image+Geo denote and how the counts were computed.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: external benchmark data with standard metrics; label incompleteness is a validity concern, not a circular derivation.

full rationale

ZooplanktonBench is a dataset and benchmark paper rather than a derivation chain. The central contribution is an externally collected image set (ISIIS tow on 24 July 2011 in the northern Gulf of Mexico) with 285,733 human-annotated organism crops and a set of detection, classification, and tracking evaluations. The reported YOLOv8, Grounding DINO, and GPT-4V numbers are computed with standard mAP metrics on held-out test images, so they measure model behavior on new data rather than restating the training labels. The paper's self-citations (e.g., refs [7]–[9], involving co-author Greer) support the imaging system and prior ecological context; none is invoked as an authority to force a result or to forbid alternative choices. The acknowledged incompleteness in Section 3 — 'smaller and more abundant organisms, such as copepods and appendicularians, were present in almost every image but were not identified' — means that the binary living-versus-marine-snow evaluation in Section 4.1.2 treats unlabeled living organisms as background. This is a label-completeness and construct-validity problem for interpreting mAP50 = 0.871 as true living-versus-non-living performance, but it is not circularity: no fitted parameter is renamed as a prediction, no result is defined in terms of itself, and the metric is computed from the labels as released. The paper itself flags the missing benchtop-video annotation as a limitation. I therefore find no significant circularity.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central benchmark rests on several unverified assumptions, most importantly the completeness of the negative class in the binary task. The paper also uses hand-chosen thresholds and an unspecified image subset for the GPT-4V evaluation.

free parameters (3)
  • GPT-4V evaluation subset size = 2,188 images
    The paper selects 2,188 representative zooplankton images for the GPT-4V experiment without describing the sampling procedure, which affects the reported accuracies.
  • Grounding DINO confidence thresholds = 0.05, 0.075, 0.1
    The open-set detector results depend strongly on the chosen threshold; the paper does not state a principled way to select it.
  • ByteTrack track length threshold = 5 frames
    In the images-plus-video track, an object is considered living only if tracked for at least 5 frames; this threshold is chosen by hand and directly changes the filtering outcome.
assumptions (3)
  • domain assumption All unlabeled image regions are marine snow (negative class) for the living-versus-marine-snow detection task.
    Section 3 notes that many living zooplankton, such as copepods and appendicularians, were not labeled, so this assumption is violated.
  • ad hoc to paper Tracking consistency across 5 frames is a reliable indicator that an object is a living zooplankton.
    Used in Section 5.1 to filter detections in the images-plus-video track; no evidence beyond a heuristic is provided.
  • domain assumption The 2,188-image subset used for GPT-4V is representative of the full dataset.
    The selection procedure is not described, but the paper draws conclusions about GPT-4V performance from this subset.

how reviews work

0 comments
Cite this review

Pith. "Pith review of ZooplanktonBench: A Geo-Aware Zooplankton Recognition and Classification Dataset from Marine Observations." pith.science (2026). https://pith.science/paper/6I35572T

@misc{pith2026250518477,
  author       = {Pith},
  title        = {Pith review of: ZooplanktonBench: A Geo-Aware Zooplankton Recognition and Classification Dataset from Marine Observations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6I35572T}},
  note         = {Machine review of arXiv:2505.18477}
}
read the original abstract

Plankton are small drifting organisms found throughout the world's oceans and can be indicators of ocean health. One component of this plankton community is the zooplankton, which includes gelatinous animals and crustaceans (e.g. shrimp), as well as the early life stages (i.e., eggs and larvae) of many commercially important fishes. Being able to monitor zooplankton abundances accurately and understand how populations change in relation to ocean conditions is invaluable to marine science research, with important implications for future marine seafood productivity. While new imaging technologies generate massive amounts of video data of zooplankton, analyzing them using general-purpose computer vision tools turns out to be highly challenging due to the high similarity in appearance between the zooplankton and its background (e.g., marine snow). In this work, we present the ZooplanktonBench, a benchmark dataset containing images and videos of zooplankton associated with rich geospatial metadata (e.g., geographic coordinates, depth, etc.) in various water ecosystems. ZooplanktonBench defines a collection of tasks to detect, classify, and track zooplankton in challenging settings, including highly cluttered environments, living vs non-living classification, objects with similar shapes, and relatively small objects. Our dataset presents unique challenges and opportunities for state-of-the-art computer vision systems to evolve and improve visual understanding in dynamic environments characterized by significant variation and the need for geo-awareness. The code and settings described in this paper can be found on our website: https://lfk118.github.io/ZooplanktonBench_Webpage.

Figures

Figures reproduced from arXiv: 2505.18477 by the authors.

Figure 1
Figure 1. (a) The In Situ Ichthyoplankton Imaging System [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The region of sampling in the nGOM. Bathymetry [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
Figure 3
Figure 3. Example photos of zooplankton in our Zooplank [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Examples of zooplankton and marine snow. Marine [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Image examples from 10 meters, 25 meters, and 35 [PITH_FULL_IMAGE:figures/full_fig_p005_5.png]
Figure 6
Figure 6. Figure 6: YOLOv8 detection and classification results. Top: [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

39 extracted references · 32 canonical work pages

  1. [1]

    We Could Run Out of Data to Train AI Language Programs.MIT Technology Review (2022)

    2022. We Could Run Out of Data to Train AI Language Programs.MIT Technology Review (2022)

    work page 2022

  2. [2]

    Mark C Benfield, Philippe Grosjean, Phil F Culverhouse, Xabier Irigoien, Michael E Sieracki, Angel Lopez-Urrutia, Hans G Dam, Qiao Hu, Cabell S Davis, Allen Hansen, et al. 2007. RAPID: research on automated plankton identification. Oceanography 20, 2 (2007), 172–187

    work page 2007

  3. [3]

    David Berthelot, Nicholas Carlini, Ian Goodfellow, Nicolas Papernot, Avital Oliver, and Colin A Raffel. 2019. Mixmatch: A holistic approach to semi-supervised learning. Advances in neural information processing systems 32 (2019)

    2019

  4. [4]

    Elijah Cole, Grant Van Horn, Christian Lange, Alexander Shepard, Patrick Leary, Pietro Perona, Scott Loarie, and Oisin Mac Aodha. 2023. Spatial implicit neural representations for global-scale species mapping. In International Conference on Machine Learning. PMLR, 6320–6342

    work page 2023

  5. [5]

    Robert K Cowen and Cedric M Guigand. 2008. In situ ichthyoplankton imaging system (ISIIS): system design and preliminary results. Limnology and Oceanogra- phy: Methods 6, 2 (2008), 126–132

    work page 2008

  6. [6]

    Jeffrey S Ellen, Casey A Graff, and Mark D Ohman. 2019. Improving plankton image classification using context metadata. Limnology and Oceanography: Methods 17, 8 (2019), 439–461

    work page 2019

  7. [7]

    Adam T Greer, Luciano M Chiaverano, Laura M Treible, Christian Briseño-Avena, and Frank J Hernandez. 2021. From spatial pattern to ecological process through imaging zooplankton interactions. ICES Journal of Marine Science 78, 8 (2021), 2664–2674

    work page 2021

  8. [8]

    Adam T Greer, Patrick I Duffy, Tim J W Walles, Charles Cousin, Laura M Treible, Kyle D Aaron, and Jens C Nejstgaard. 2025. Modular shadowgraph imaging for zooplankton ecological studies in diverse field and mesocosm settings. Lim- nology and Oceanography: Methods 23, 1 (2025), 67–86. doi:10.1002/lom3.10657 arXiv:https://aslopubs.onlinelibrary.wiley.com/do...

    work page doi:10.1002/lom3.10657 2025

Show all 39 references
  1. [9]

    Adam T Greer, C Brock Woodson, Conner E Smith, Cedric M Guigand, and Robert K Cowen. 2016. Examining mesozooplankton patch structure and its implications for trophic interactions in the northern Gulf of Mexico. Journal of Plankton Research 38, 4 (2016), 1115–1134

    2016

  2. [10]

    Churchill B Grimes. 2001. Fishery production and the Mississippi River discharge. Fisheries 26, 8 (2001), 17–26

    2001

  3. [11]

    Derek Hoiem, Yodsawalai Chodpathumwan, and Qieyun Dai. 2012. Diagnosing error in object detectors. In European conference on computer vision . Springer, 340–353

    2012

  4. [12]

    Qiao Hu and Cabell Davis. 2005. Automatic plankton image recognition with co-occurrence matrices and support vector machine. Marine Ecology Progress Series 295 (2005), 21–31

    2005

  5. [13]

    Glenn Jocher, Ayush Chaurasia, and Jing Qiu. 2023. Ultralytics YOLO. https: //github.com/ultralytics/ultralytics

    2023

  6. [14]

    Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C Lawrence Zitnick. 2014. Microsoft coco: Common objects in context. In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proce...

    2014

  7. [15]

    Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023. Visual Instruc- tion Tuning

    2023

  8. [16]

    Li Liu, Wanli Ouyang, Xiaogang Wang, Paul Fieguth, Jie Chen, Xinwang Liu, and Matti Pietikäinen. 2020. Deep learning for generic object detection: A survey. International journal of computer vision 128 (2020), 261–318

    2020

  9. [17]

    Shilong Liu, Zhaoyang Zeng, Tianhe Ren, Feng Li, Hao Zhang, Jie Yang, Chunyuan Li, Jianwei Yang, Hang Su, Jun Zhu, et al . 2023. Grounding dino: Marrying dino with grounded pre-training for open-set object detection. arXiv preprint arXiv:2303.05499 (2023)

    2023 arXiv

  10. [18]

    Fabien Lombard, Emmanuel Boss, Anya M Waite, Meike Vogt, Julia Uitz, Lars Stemmann, Heidi M Sosik, Jan Schulz, Jean-Baptiste Romagnan, Marc Picheral, et al. 2019. Globally consistent quantitative observations of planktonic ecosys- tems. Frontiers in Marine Science 6 (2019), 196

    2019

  11. [19]

    Haitong Lou, Xuehu Duan, Junmei Guo, Haiying Liu, Jason Gu, Lingyun Bi, and Haonan Chen. 2023. DC-YOLOv8: small-size object detection algorithm based on camera sensor. Electronics 12, 10 (2023), 2323

    2023

  12. [20]

    Jessica Y Luo, Jean-Olivier Irisson, Benjamin Graham, Cedric Guigand, Amin Sarafraz, Christopher Mader, and Robert K Cowen. 2018. Automated plankton im- age analysis using convolutional neural networks. Limnology and Oceanography: methods 16, 12 (2018), 814–827

    2018

  13. [21]

    Oisin Mac Aodha, Elijah Cole, and Pietro Perona. 2019. Presence-only geograph- ical priors for fine-grained image classification. In Proceedings of the IEEE/CVF International Conference on Computer Vision . 9596–9606

    2019

  14. [22]

    Gengchen Mai, Krzysztof Janowicz, Bo Yan, Rui Zhu, Ling Cai, and Ni Lao. 2020. Multi-Scale Representation Learning for Spatial Feature Distributions using Grid Cells. In ICLR 2020. openreview

    2020

  15. [23]

    Gengchen Mai, Ni Lao, Yutong He, Jiaming Song, and Stefano Ermon. 2023. Csp: Self-supervised contrastive spatial pre-training for geospatial-visual representa- tions. In International Conference on Machine Learning . PMLR, 23498–23515

    2023

  16. [24]

    Gengchen Mai, Yao Xuan, Wenyun Zuo, Yutong He, Jiaming Song, Stefano Er- mon, Krzysztof Janowicz, and Ni Lao. 2023. Sphere2Vec: A general-purpose location representation learning over a spherical surface for large-scale geospa- tial predictions. ISPRS Journal of Photogrammetry...

    2023

  17. [25]

    National Oceanic and Atmospheric Administration. 2020. What is marine snow? - National Ocean Service. https://oceanservice.noaa.gov/facts/marinesnow.html#: ~:text=Marine%20snow%20is%20a%20shower,finally%20reaching%20the% 20ocean%20floor

    2020

  18. [26]

    OpenAI. 2024. ChatGPT (GPT-4V). https://www.openai.com

    2024

  19. [27]

    Nikhil Parthasarathy, SM Eslami, Joao Carreira, and Olivier Henaff. 2023. Self- supervised video pretraining yields robust and more human-aligned visual repre- sentations. Advances in Neural Information Processing Systems 36 (2023), 65743– 65765

    2023

  20. [28]

    Alec Radford, Jeff Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language Models are Unsupervised Multitask Learners. (2019)

    2019

  21. [29]

    Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi. 2016. You only look once: Unified, real-time object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition . 779–788

    2016

  22. [30]

    Kelly L Robinson, Su Sponaugle, Jessica Y Luo, Miram R Gleiber, and Robert K Cowen. 2021. Big or small, patchy all: Resolution of marine plankton patch structure at micro-to submesoscales for 36 taxa. Science Advances 7, 47 (2021), eabk2904

    2021

  23. [31]

    Moritz S Schmid, Dominic Daprano, Malhar M Damle, Christopher M Sullivan, Su Sponaugle, Charles Cousin, Cedric Guigand, and Robert K Cowen. 2023. Edge computing at sea: High-throughput classification of in-situ plankton imagery for adaptive sampling. Frontiers in Marine Scienc...

    2023

  24. [32]

    Caroline A Schneider, Wayne S Rasband, and Kevin W Eliceiri. 2012. NIH Image to ImageJ: 25 years of image analysis. Nature methods 9, 7 (2012), 671–675

    2012

  25. [33]

    Heidi M Sosik and Robert J Olson. 2007. Automated taxonomic classification of phytoplankton sampled with imaging-in-flow cytometry. Limnology and Oceanography: Methods 5, 6 (2007), 204–216

    2007

  26. [34]

    Fatma M Talaat and Hanaa ZainEldin. 2023. An improved fire detection approach based on YOLO-v8 for smart cities. Neural Computing and Applications 35, 28 (2023), 20939–20954

    2023

  27. [35]

    Jefferson T Turner. 2015. Zooplankton fecal pellets, marine snow, phytodetritus and the ocean’s biological pump. Progress in Oceanography 130 (2015), 205–248

    2015

  28. [36]

    Nemin Wu, Qian Cao, Zhangyu Wang, Zeping Liu, Yanlin Qi, Jielu Zhang, Joshua Ni, X Angela Yao, Hongxu Ma, Lan Mu, et al . 2024. TorchSpatial: A Location Encoding Framework and Benchmark for Spatial Representation Learning. In The Thirty-eight Conference on Neural Information P...

    2024

  29. [37]

    Bingjie Xiao, Minh Nguyen, and Wei Qi Yan. 2024. Fruit ripeness identification using YOLOv8 model. Multimedia Tools and Applications 83, 9 (2024), 28039– 28056

    2024

  30. [38]

    Lingfeng Yang, Xiang Li, Renjie Song, Borui Zhao, Juntian Tao, Shihao Zhou, Jiajun Liang, and Jian Yang. 2022. Dynamic mlp for fine-grained image classifica- tion by leveraging geographical and temporal information. In Proceedings of the IEEE/CVF conference on computer vision ...

    2022

  31. [39]

    Yifu Zhang, Peize Sun, Yi Jiang, Dongdong Yu, Fucheng Weng, Zehuan Yuan, Ping Luo, Wenyu Liu, and Xinggang Wang. 2022. ByteTrack: Multi-Object Tracking by Associating Every Detection Box. arXiv:2110.06864 [cs.CV]

    2022 arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.