Leveraging Co‐Occurrence to Improve Deep Learning Photo‐Identification in Social Animals

Barnhill A, Towers JR, Sutton GJ, Shaw TJ, Doniol-Valcroze T, Maier A, Nöth E, Bergler C (2026)

Publication Type: Journal article

Publication year: 2026

Journal

Ecology and Evolution Wiley Open Access

Original Authors: Alexander Barnhill, Jared R. Towers, Gary J. Sutton, Tasli J. H. Shaw, Thomas Doniol‐Valcroze, Andreas Maier, Elmar Nöth, Christian Bergler

Book Volume: 16

Article Number: e73552

Issue: 4

Journal Issue: 4

DOI: 10.1002/ece3.73552

Abstract

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.

Authors with CRIS profile

Alexander Barnhill Lehrstuhl für Informatik 14 (Bild- und Sprachverarbeitung) (LME) Andreas Maier Lehrstuhl für Informatik 5 (Mustererkennung) Elmar Nöth Professur für Informatik (Mustererkennung) Christian Bergler Lehrstuhl für Informatik 5 (Mustererkennung)

Involved external institutions

Bay Cetology CA Canada (CA) Fisheries and Oceans Canada / Department of Fisheries and Oceans (DFO) / Pêches et Océans Canada CA Canada (CA)

How to cite

APA:

Barnhill, A., Towers, J.R., Sutton, G.J., Shaw, T.J., Doniol-Valcroze, T., Maier, A.,... 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

MLA:

Barnhill, Alexander, et al. "Leveraging Co‐Occurrence to Improve Deep Learning Photo‐Identification in Social Animals." Ecology and Evolution 16.4 (2026).

BibTeX: Download