An Integer Programming Approach to Re-identification in 1 Livestock Tracking with Quantum-Ready Adaptation

Accurate tracking of individual animals is essential for precision livestock farming, enabling health monitoring, behavior analysis, and phenotyping. Multi-object tracking in livestock environments is challenging due to occlusions, frequent interactions, and similar appearances among individuals. In this study, we present FlowReID, a post-processing re-identification (reID) approach that extends the BoT-SORT tracking framework such that is suitable for optimization through quantum com11 puting. For this, we formulate the problem of reID as a binary linear program, which we then reformulate as a Quadratic Unconstrained Binary Optimization (QUBO) to allow for optimization through quantum annealing. Evaluation on video sequences of 16 Holstein Friesian cows from multi ple camera views demonstrated that FlowReID consistently improved identity consistency compared to the standard Ultralytics reID module, reducing the number of identity switches from 9 to 5 while slightly improving overall multi-object tracking metrics like HOTA (89.7 to 90.8), MOTA (95.4 to 96.3), and IDF1 (96.0 to 97.4). Although our benchmarks indicate that quantum computing solvers were slower and less accurate than traditional approaches, one should note that the development of quantum computers is still in its infancy, and rapid improvements in quantum hardware are ex20 pected in the near future, on which our developed quantum algorithm FlowReID should be directly applicable. Thereby providing a proof-of-concept for integrating quantum computing into real-world livestock tracking applications, illustrating how an existing tracking problem can be transformed into a form suitable for quantum computing. FlowReID demonstrates both practical benefits for current precision livestock applications and a foundation for future exploration of scalable, high-complexity tracking systems.