Are We Recognizing the Jaguar or Its Background? A Diagnostic Framework for Jaguar Re-Identification

TL;DR

This paper presents a diagnostic framework for wildlife re-identification, emphasizing the importance of understanding what visual evidence models rely on, and introduces a new benchmark for jaguar re-ID.

Contribution

It introduces a diagnostic framework with novel metrics and a curated jaguar benchmark to evaluate whether models use correct visual cues for re-identification.

Findings

Models often rely on background or silhouette rather than coat patterns.

The diagnostic framework reveals the true basis of model decisions.

Mitigation strategies can shift model reliance towards correct features.

Abstract

Jaguar re-identification (re-ID) from citizen-science imagery can look strong on standard retrieval metrics while still relying on the wrong evidence, such as background context or silhouette shape, instead of the coat pattern that defines identity. We introduce a diagnostic framework for wildlife re-ID with two axes: a leakage-controlled context ratio, background/foreground, computed from inpainted background-only versus foreground-only images, and a laterality diagnostic based on cross-flank retrieval and mirror self-similarity. To make these diagnostics measurable, we curate a Pantanal jaguar benchmark with per-pixel segmentation masks and an identity-balanced evaluation protocol. We then use representative mitigation families, ArcFace fine-tuning, anti-symmetry regularization, and Lorentz hyperbolic embeddings, as case studies under the same evaluation lens. The goal is not only to ask which model ranks best, but also what visual evidence it uses to do so.