WildFusion: Individual Animal Identification with Calibrated Similarity Fusion

TL;DR

WildFusion is a novel method that combines deep global descriptors and local matching similarities, calibrated for improved accuracy in individual animal identification across diverse species and datasets.

Contribution

It introduces a new fusion and calibration approach for animal identification, outperforming existing models in zero-shot and dataset-specific settings.

Findings

Achieved 76.2% accuracy using local similarity scores alone in zero-shot settings.

Improved accuracy to 84.0% with combined global and local scores and calibration.

Significantly reduced mean relative error by 35%.

Abstract

We propose a new method - WildFusion - for individual identification of a broad range of animal species. The method fuses deep scores (e.g., MegaDescriptor or DINOv2) and local matching similarity (e.g., LoFTR and LightGlue) to identify individual animals. The global and local information fusion is facilitated by similarity score calibration. In a zero-shot setting, relying on local similarity score only, WildFusion achieved mean accuracy, measured on 17 datasets, of 76.2%. This is better than the state-of-the-art model, MegaDescriptor-L, whose training set included 15 of the 17 datasets. If a dataset-specific calibration is applied, mean accuracy increases by 2.3% percentage points. WildFusion, with both local and global similarity scores, outperforms the state-of-the-art significantly - mean accuracy reached 84.0%, an increase of 8.5 percentage points; the mean relative error drops by 35%. We make the code and pre-trained models publicly available5, enabling immediate use in ecology and conservation.