Fine-tuned YOLOv26m object detector on the Brackish Underwater benchmark dataset, trained and evaluated as part of DetectionBench -- a framework for reproducibly benchmarking modern object detectors with identical training recipes and evaluation metrics across multiple real-world datasets.
Metrics reported in this model card are computed on the Brackish Underwater test split, using DetectionBench's standard evaluation pipeline (detectionbench-evaluate).
Brackish Underwater Model Zoo
Every model DetectionBench has trained and evaluated on Brackish Underwater so far, for full transparency -- see DetectionBench for the smaller, curated comparison set used on the project README.
Model
mAP@50
mAP@50-95
Precision
Recall
YOLOv8s
99.3
85.65
99.36
98.53
YOLOv8m
99.19
86.04
98.69
98.71
YOLOv26s
99.1
85.77
99.45
97.83
YOLOv26n
98.95
83.66
98.17
97.23
YOLOv11x
98.89
86.3
98.84
98.25
YOLOv11n
98.87
83.27
98.76
96.69
YOLOv26m
98.74
85.93
98.74
97.32
YOLOv8n
98.41
83.01
99.51
96.95
Per-Class Performance
Class
mAP@50
mAP@50-95
crab
99.5
95.53
fish
99.44
89.89
jellyfish
96.31
73.13
shrimp
99.47
79.23
small_fish
98.21
78.42
starfish
99.5
99.4
Evaluation Visualizations
Precision-Recall Curve
PR Curve
F1 Curve
F1 Curve
Confusion Matrix
Confusion Matrix
Dataset
This model was trained on Brackish Underwater. For the full dataset description, provenance, license, and citation, see the dataset card:
DetectionBench -- reproducible benchmarks for modern object detectors on real-world datasets
Training Framework
This model was trained using DetectionBench, an open-source framework for benchmarking object detectors across multiple real-world datasets with a common pipeline.
Features include:
A dataset-adapter registry for converting real-world datasets into a canonical format
Identical training/evaluation recipes across model families (Ultralytics YOLO/RT-DETR, RF-DETR)
One-command reproducibility via versioned Hydra configs
If you find this model useful, please consider starring the repository.
Known Limitations
Severe class imbalance: crab (34.6%) and small_fish (30.4%) account for roughly two-thirds of all annotated boxes in the training set, while shrimp (1.46%) and jellyfish (1.82%) are rare -- per-class accuracy on the minority classes is measured on comparatively few examples.
Roughly 15% of images have no annotated objects at all (background-only frames, by design -- part of the dataset's varying-visibility setup, not a data-quality issue).
Single-location, single-camera capture: all footage comes from one fixed camera 9 meters below the surface on the Limfjords bridge, Denmark -- generalization to other underwater cameras, locations, or water types (brackish vs. marine/freshwater) is untested.
Two-hop provenance: this dataset was converted to YOLO format via a third-party Roboflow export, not sourced directly from the original annotated-video release; images are pre-resized to 1920x1080 by that export.
Citation
If you use this model in your research, please consider citing:
The Brackish Underwater dataset (see below)
The original YOLOv26m architecture (see below)
The other model architectures shown in the Model Zoo/External Comparison tables above, if you reference their results
DetectionBench, the training/evaluation framework used to produce this checkpoint
@InProceedings{pedersen2019brackish,
title = {Detection of Marine Animals in a New Underwater Dataset with Varying Visibility},
author = {Pedersen, Malte and Haurum, Joakim Bruslund and Gade, Rikke and Moeslund, Thomas B. and Madsen, Niels},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
month = {June},
year = {2019}
}
bibtex
1@article{jocher2026yolo26,
2 title={Ultralytics YOLO26: Unified Real-Time End-to-End Vision Models},
3 author={Jocher, Glenn and Qiu, Jing and Liu, Mengyu and Lyu, Shuai and Akyon, Fatih Cagatay and Kalfaoglu, Muhammet Esat},
4 journal={arXiv preprint arXiv:2606.03748},
5 year={2026}
6}
Other architectures compared against on Brackish Underwater in this model card:
YOLOv11
bibtex
1No official YOLO11 research paper has been published by Ultralytics; the most commonly cited independent architectural analysis is used instead:
23@article{khanam2024yolov11,
4 title={YOLOv11: An Overview of the Key Architectural Enhancements},
5 author={Khanam, Rahima and Hussain, Muhammad},
6 journal={arXiv preprint arXiv:2410.17725},
7 year={2024}
8}
YOLOv8
bibtex
1No official YOLOv8 research paper has been published by Ultralytics; this is their own recommended software citation instead:
23@software{jocher2023yolov8,
4 author = {Glenn Jocher and Ayush Chaurasia and Jing Qiu},
5 title = {Ultralytics YOLOv8},
6 version = {8.0.0},
7 year = {2023},
8 url = {https://github.com/ultralytics/ultralytics},
9 license = {AGPL-3.0}
10}
bibtex
1@software{Saksena_DetectionBench_2026,
2 author = {Saksena, Saumya Kumaar},
3 title = {DetectionBench: Reproducible Benchmarks for Modern Object Detectors on Real-World Datasets},
4 url = {https://github.com/dronefreak/DetectionBench},
5 year = {2026}
6}
