Automated detection of Antarctic benthic organisms in high-resolution in situ imagery to aid biodiversity monitoring: optimal model weights
GB/NERC/BAS/PDC/02070
Abstract:
Model weights for the optimal object detection model, trained on the Weddell Sea Benthic Dataset. Trained 2025-05. Weights should be used with a Deformable-DETR architecture.
This work was funded by the UKRI Future Leaders Fellowship MR/W01002X/1 'The past, present and future of unique cold-water benthic (sea floor) ecosystems in the Southern Ocean' awarded to Rowan Whittle.
Citation:
Trotter, C., Griffiths, H.J., Khan, T.M., & Whittle, R.J. (2025). Automated detection of Antarctic benthic organisms in high-resolution in situ imagery to aid biodiversity monitoring: optimal model weights (Version 1.0) [Data set]. NERC EDS UK Polar Data Centre. https://doi.org/10.5285/b2874f3f-285d-4ae6-9bb4-6bfe3eacbfff
Access Data
GET DATA
PROJECT HOME PAGE
REFERENCE MATERIALS
- https://doi.org/10.1109/ICIP46576.2022.9897990
- https://doi.org/10.3390/info11020125
- https://doi.org/10.48550/arXiv.1906.07155
- https://doi.org/10.48550/arXiv.2010.04159
RELATED DATA SET METADATA
SOFTWARE PACKAGES
License
| Access: | None |
|---|---|
| Use: | Data are supplied under Open Government Licence v3.0 http://www.nationalarchives.gov.uk/doc/open-government-licence/version/3/. |
Basic Information
| Creation Date: | 2025-06-06 |
|---|---|
| Dataset Progress: | Complete |
| Dataset Language: | English |
| ISO Topic Categories: |
|
| Parameters: |
|
Additional Information
| Reference: | Associated publication: Trotter, C., Griffiths, H., Khan, T.M., Whittle, R., 2025. Automated Detection of Antarctic Benthic Organisms in High-Resolution In Situ Imagery to Aid Biodiversity Monitoring, in: 2025 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW), IEEE, Honolulu, HI, USA (in preparation) References: [1] Zhu, X., Su, W., Lu, L., Li, B., Wang, X., Dai, J., 2021. Deformable DETR: Deformable Transformers for End-to-End Object Detection. https://doi.org/10.48550/arXiv.2010.04159 [2] Chen, K., Wang, Jiaqi, Pang, J., Cao, Y., Xiong, Y., Li, X., Sun, S., Feng, W., Liu, Z., Xu, J., Zhang, Z., Cheng, D., Zhu, C., Cheng, T., Zhao, Q., Li, B., Lu, X., Zhu, R., Wu, Y., Dai, J., Wang, Jingdong, Shi, J., Ouyang, W., Loy, C.C., Lin, D., 2019. MMDetection: Open MMLab Detection Toolbox and Benchmark. https://doi.org/10.48550/arXiv.1906.07155 [3] Akyon, F.C., Onur Altinuc, S., Temizel, A., 2022. Slicing Aided Hyper Inference and Fine-Tuning for Small Object Detection, in: 2022 IEEE International Conference on Image Processing (ICIP). Presented at the 2022 IEEE International Conference on Image Processing (ICIP), IEEE, Bordeaux, France, pp. 966-970. https://doi.org/10.1109/ICIP46576.2022.9897990 [4] Buslaev, A., Iglovikov, V.I., Khvedchenya, E., Parinov, A., Druzhinin, M., Kalinin, A.A., 2020. Albumentations: Fast and Flexible Image Augmentations. Information 11, 125. https://doi.org/10.3390/info11020125 |
|
|---|---|---|
| Lineage/Methodology: | Model weights for a Deformable-DETR architecture, trained to detect 25 Antarctic benthic morphotypes using the Weddell Sea Benthic Dataset (see Related URLs). Model version: 1.0.0. Model weights correspond to the optimal object detection model capable of automated benthic organism detection as described in [in-prep]. For model development code, (see Related URLs). |
|
Extent
| Temporal Coverage: | |
|---|---|
| Start Date | 2025-05-28 |
| End Date | N/A |
Instrumentation
| Data Collection: | The model was trained using a single High Performance Computing node with CentOS 7, and one NVIDIA A2 GPU. Training was performed using Python v3.8.20 alongside the following packages: MMDetection v3.3.0 [1] Pytorch v2.4.0 Pytorch-cuda v12.1 torchvision v0.19.0 sahi v0.11.22 [2] albumentations v1.3.1 [3] |
|---|
Storage
| Data Storage: | Pytorch model weights: 1 file (pth format), 164.1MB |
|---|