Leveraging Co‐Occurrence to Improve Deep Learning Photo‐Identification in Social Animals
Ecology and Evolution · 2026-04 · 16 (4)
DOI: 10.1002/ece3.73552 ↗At a glance
Journal article in Ecology and Evolution, published online on 2026-04-21. Written by Alexander Barnhill and 7 co-authors. Institution: Friedrich-Alexander-Universität Erlangen-Nürnberg Pattern Recognition Lab Erlangen Germany. Cited 1 times so far (Crossref); 54 references. The full text is freely available (open access, CC BY 4.0).
Abstract (original language)
Photo‐identification underpins individual‐based inference in numerous ecological studies, but scaling it to decades‐long archives remains limited by expert time. Deep learning can accelerate matching, yet most pipelines treat photographs as independent observations and therefore ignore a key aspect of the data collection method: individuals are recorded in structured encounters and often exhibit persistent, non‐random associations. We present a model agnostic, encounter‐level identification procedure that incorporates social context as a deployable probabilistic component. Given per‐image classifier posteriors, we perform log‐linear fusion of three information sources: (i) image‐based probabilities, (ii) global sighting priors (class frequency), and (iii) an encounter‐conditioned context term derived from historical co‐occurrence (log lift). The method operates as lightweight post‐processing and requires no retraining or architectural changes to the image model. Using a longitudinal photo‐identification dataset as a case study (West Coast Transient Bigg's killer whales), we evaluate (a) expert‐assisted settings in which a small number of individuals present in an encounter are known without image‐level labels, and (b) fully automated settings that initialize context from the model's own high‐confidence predictions. On a strict temporal holdout (newest 10%), encounter‐context fusion reduces top‐1 error by ~14%–25% with expert‐assisted seeding; a fully automated variant yields up to ~24% fewer misidentifications once sufficient training history exists, improving Macro‐F1 by +0.088 to +0.104, with minimal computational overhead. Placebo and seed‐corruption controls confirm that gains depend on meaningful co‐occurrence structure and collapse when encounter context is destroyed. By turning encounter structure into a reusable probabilistic component, this work bridges established methods for analyzing animal societies with practical, scalable photo‐identification pipelines. The approach is applicable to any system where individuals are repeatedly observed in groups (e.g., cetaceans, primates, ungulates, camera‐trap bursts) and provides a transparent mechanism to incorporate social context into automated identification.
Subjects
Author keywords
Authors and institutions
- Friedrich-Alexander-Universität Erlangen-Nürnberg Pattern Recognition Lab Erlangen Germany.
- Bay Cetology Alert Bay British Columbia Canada.
- Pacific Biological Station, Fisheries and Oceans Canada Nanaimo British Columbia Canada.
- Ocean Wise Vancouver British Columbia Canada.
- Department of Electrical Engineering, Media and Computer Science Ostbayerische Technische Hochschule Amberg-Weiden Amberg Germany.
ORCID profiles: Alexander Barnhill, Jared R Towers
Cite
APA
Barnhill, A., Towers, J. R., Sutton, G. J., Shaw, T. J. H., Doniol-Valcroze, T., Maier, A., Nöth, E., & Bergler, C. (2026). Leveraging Co‐Occurrence to Improve Deep Learning Photo‐Identification in Social Animals. Ecology and Evolution, 16(4). https://doi.org/10.1002/ece3.73552
Vancouver
Barnhill A, Towers JR, Sutton GJ, Shaw TJH, Doniol-Valcroze T, Maier A, et al. Leveraging Co‐Occurrence to Improve Deep Learning Photo‐Identification in Social Animals. Ecology and Evolution. 2026;16(4). doi:10.1002/ece3.73552
BibTeX
@article{Barnhill2026,
author = {Barnhill, Alexander and Towers, Jared R and Sutton, Gary J and Shaw, Tasli J H and Doniol-Valcroze, Thomas and Maier, Andreas and Nöth, Elmar and Bergler, Christian},
title = {Leveraging Co‐Occurrence to Improve Deep Learning Photo‐Identification in Social Animals},
journal = {Ecology and Evolution},
year = {2026},
volume = {16},
number = {4},
doi = {10.1002/ece3.73552}
}
EUVIMED