Taxonomy-aware deep learning for hierarchical marine species classification in underwater imagery
For marine ecologists and conservationists needing scalable biodiversity monitoring, this work addresses domain shift and uneven annotation granularity in underwater image classification.
A taxonomy-aware deep learning framework for hierarchical marine species classification achieves a mean taxonomic distance of 1.581 on the FathomNet 2025 dataset, within 3% of the 1st-place solution (1.535), with gains from metric-aligned inference and decoupled components that generalize better under distribution shift.
Automated classification of marine species from underwater imagery is essential for scalable ocean biodiversity monitoring and conservation policy. Existing approaches struggle with severe domain shift across collection platforms, fine-grained visual similarity between closely related species, and uneven annotation granularity, where many specimens can only be identified to genus or a coarser taxonomic rank. We present a taxonomy-aware deep learning framework that aligns both the training loss and the inference rule with the hierarchical structure of biological classification, combining a taxonomy-weighted loss, minimum-risk Bayesian inference, multi-scale feature encoding, and independent per-rank classification heads. Evaluated on the FathomNet 2025 dataset1 (79 marine classes across seven taxonomic ranks), the system achieves a mean taxonomic distance of 1.581, within 3% of the 1st-place solution (1.535), with the largest gains from metric-aligned inference and simple, decoupled components that generalize better than learned dependencies under distribution shift.