ReCowGnition: A Realistic Biometric Benchmark for Cow Face Recognition

2026-07-24

AI Summary

This study addresses the challenges of cattle facial recognition in real-world farm settings—such as high visual similarity among individuals, frequent occlusions, non-cooperative behaviors, and scarce annotated data—by introducing the first large-scale, publicly available dairy cow facial dataset captured under authentic farming conditions. The dataset comprises 6,838 images from 161 individual cows and includes two verification and four identification evaluation protocols. Leveraging this benchmark, the authors systematically evaluate six representative approaches, including limited-data training, cross-species transfer learning, and zero-shot foundation models, thereby establishing robust performance baselines. This work provides a reproducible and comparable experimental framework, along with essential data resources, to advance research in bovine facial recognition.

Abstract

With the development of precision livestock farming and the advances in computer vision, visual animal biometrics has gained attention. Using biometric technologies that have been proven effective for humans to identify livestock can increase animal welfare as well as production efficiency. However, challenges such as complex scenarios, similar appearances, occlusions, and non-cooperative behavior, as well as the limited amount of publicly available labeled datasets, remain. In this work, we contribute a novel, publicly available cow face benchmark dataset that has been collected in a realistic automatic scenario with 6,838 images of 161 different cows at a dairy farm. In addition to the public dataset, we define two verification and four identification evaluation protocols to foster comparable research in the cow recognition research field. Further, we provide evaluation results on our dataset of six benchmark models, which include models trained on limited data, cross-species fine-tuned models, and zero-shot foundation model approaches.

Problem

• cow face recognition
• biometric benchmark
• realistic scenario
• limited dataset
• animal biometrics

Innovation

• cow face recognition
• biometric benchmark
• realistic dataset
• evaluation protocols
• foundation models