Benchmarking pig detection and tracking under diverse and challenging conditions

2025-07-22

AI Summary

A lack of systematic benchmarking for detection and tracking algorithms in individual pig behavior monitoring hinders progress in intelligent swine farming. Method: We introduce PigDetect (for detection) and PigTrack (for tracking), the first high-quality, real-world pig-farm datasets, and conduct the first end-to-end, cross-environment benchmark evaluation. We comprehensively assess YOLO-based detectors, end-to-end trainable trackers, and SORT-family association methods, incorporating challenging sample sampling to enhance robustness. Contribution/Results: SORT-based approaches achieve the best trade-off between accuracy and efficiency. Models trained on our datasets demonstrate strong generalization to unseen pens. Both datasets and code are publicly released, establishing a reproducible benchmark and technical foundation for animal welfare monitoring and automated management in precision swine farming.

Technology Category

• Computer Vision: Motion & Tracking
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

To ensure animal welfare and effective management in pig farming, monitoring individual behavior is a crucial prerequisite. While monitoring tasks have traditionally been carried out manually, advances in machine learning have made it possible to collect individualized information in an increasingly automated way. Central to these methods is the localization of animals across space (object detection) and time (multi-object tracking). Despite extensive research of these two tasks in pig farming, a systematic benchmarking study has not yet been conducted. In this work, we address this gap by curating two datasets: PigDetect for object detection and PigTrack for multi-object tracking. The datasets are based on diverse image and video material from realistic barn conditions, and include challenging scenarios such as occlusions or bad visibility. For object detection, we show that challenging training images improve detection performance beyond what is achievable with randomly sampled images alone. Comparing different approaches, we found that state-of-the-art models offer substantial improvements in detection quality over real-time alternatives. For multi-object tracking, we observed that SORT-based methods achieve superior detection performance compared to end-to-end trainable models. However, end-to-end models show better association performance, suggesting they could become strong alternatives in the future. We also investigate characteristic failure cases of end-to-end models, providing guidance for future improvements. The detection and tracking models trained on our datasets perform well in unseen pens, suggesting good generalization capabilities. This highlights the importance of high-quality training data. The datasets and research code are made publicly available to facilitate reproducibility, re-use and further development.

Problem

• Benchmarking pig detection and tracking in diverse conditions
• Improving detection performance with challenging training images
• Comparing SORT-based and end-to-end models for tracking

Innovation

• Utilizes diverse datasets for pig detection and tracking
• State-of-the-art models enhance detection and tracking quality
• Public datasets and code promote reproducibility and development