GorillaWatch: An Automated System for In-the-Wild Gorilla Re-Identification and Population Monitoring

TL;DR

GorillaWatch is an automated system that combines new datasets, deep learning, and innovative techniques to improve in-the-wild gorilla re-identification and population monitoring, reducing manual effort and enhancing accuracy.

Contribution

The paper introduces GorillaWatch, a comprehensive pipeline with novel datasets, self-supervised pretraining, and explainability methods for robust gorilla re-identification in natural environments.

Findings

Large-scale datasets improve re-identification accuracy.

Self-supervised pretraining enhances domain-specific feature learning.

Feature aggregation outperforms specialized video architectures.

Abstract

Monitoring critically endangered western lowland gorillas is currently hampered by the immense manual effort required to re-identify individuals from vast archives of camera trap footage. The primary obstacle to automating this process has been the lack of large-scale, "in-the-wild" video datasets suitable for training robust deep learning models. To address this gap, we introduce a comprehensive benchmark with three novel datasets: Gorilla-SPAC-Wild, the largest video dataset for wild primate re-identification to date; Gorilla-Berlin-Zoo, for assessing cross-domain re-identification generalization; and Gorilla-SPAC-MoT, for evaluating multi-object tracking in camera trap footage. Building on these datasets, we present GorillaWatch, an end-to-end pipeline integrating detection, tracking, and re-identification. To exploit temporal information, we introduce a multi-frame self-supervised pretraining strategy that leverages consistency in tracklets to learn domain-specific features without manual labels. To ensure scientific validity, a differentiable adaptation of AttnLRP verifies that our model relies on discriminative biometric traits rather than background correlations. Extensive benchmarking subsequently demonstrates that aggregating features from large-scale image backbones outperforms specialized video architectures. Finally, we address unsupervised population counting by integrating spatiotemporal constraints into standard clustering to mitigate over-segmentation. We publicly release all code and datasets to facilitate scalable, non-invasive monitoring of endangered species