[2603.04314] MOO: A Multi-view Oriented Observations Dataset for Viewpoint Analysis in Cattle Re-Identification

Computer Science > Computer Vision and Pattern Recognition arXiv:2603.04314 (cs) [Submitted on 4 Mar 2026] Title:MOO: A Multi-view Oriented Observations Dataset for Viewpoint Analysis in Cattle Re-Identification Authors:William Grolleau, Achraf Chaouch, Astrid Sabourin, Guillaume Lapouge, Catherine Achard View a PDF of the paper titled MOO: A Multi-view Oriented Observations Dataset for Viewpoint Analysis in Cattle Re-Identification, by William Grolleau and 4 other authors View PDF HTML (experimental) Abstract:Animal re-identification (ReID) faces critical challenges due to viewpoint variations, particularly in Aerial-Ground (AG-ReID) settings where models must match individuals across drastic elevation changes. However, existing datasets lack the precise angular annotations required to systematically analyze these geometric variations. To address this, we introduce the Multi-view Oriented Observation (MOO) dataset, a large-scale synthetic AG-ReID dataset of $1,000$ cattle individuals captured from $128$ uniformly sampled viewpoints ($128,000$ annotated images). Using this controlled dataset, we quantify the influence of elevation and identify a critical elevation threshold, above which models generalize significantly better to unseen views. Finally, we validate the transferability to real-world applications in both zero-shot and supervised settings, demonstrating performance gains across four real-world cattle datasets and confirming that synthetic geometric priors effect...