A Non-Invasive Alternative to RFID: Self-Sufficient 3D Identification of Group-Housed Livestock

arXiv cs.CV / 4/27/2026

📰 NewsDeveloper Stack & InfrastructureSignals & Early TrendsModels & Research

共有:

Key Points

The paper targets a core challenge in precision livestock management: accurately identifying individual animals in group-housed settings where conventional RFID ear tags are invasive and often unreliable due to tag loss and antenna field limits.
It proposes a non-intrusive, vision-based identification system using 3D point cloud data captured inside a commercial electronic feeding station (EFS) rather than relying on RFID.
The system introduces TARA (Temporal Adaptive Recognition Architecture), a semi-supervised framework that preserves identity consistency over time through dynamic recalibration of each animal’s identity profile as morphology changes.
To reduce dependence on scarce labels, it uses visit-level majority voting to create high-quality pseudo-labels from raw temporal sequences during training.
On an operational barn dataset (group-housed sows), the method reports 100% identification accuracy at the visit level, suggesting 3D point-cloud vision could replace RFID for autonomous individual monitoring.

Abstract

Accurate identification of individual farm animals in group-housed environments is a cornerstone of precision livestock management. However, current industry standards rely heavily on Radio Frequency Identification (RFID) ear tags, which are invasive, prone to loss, and restricted by the spatial limitations of antenna fields. In this paper, we propose a non-intrusive, vision-based identification system leveraging 3D point cloud data captured within a commercial electronic feeding station (EFS). Departing from traditional supervised frame-level inference, we introduce the Temporal Adaptive Recognition Architecture (TARA), a self-sufficient, semi-supervised framework designed to maintain identity consistency over time. TARA employs a dynamic recalibration mechanism that updates individual identity profiles to account for morphological changes in the livestock. To facilitate training in label-scarce environments, we utilize a visit-level majority voting strategy to generate high-fidelity pseudo-labels from raw temporal sequences. Experimental results on a group housed sow dataset collected from an operational commercial barn demonstrate that our approach achieves 100% identification accuracy at the visit level. These results suggest that vision-based 3D point cloud analysis offers a robust, superior alternative to RFID-based systems, paving the way for fully autonomous individual animal monitoring.