Horse Eye Blink Detection and Classification for Equine Affective State Assessment

2026-06-03

AI Summary

This study addresses the challenge of manually detecting subtle and transient equine eyelid micro-expressions—such as half-blinks and full blinks—which serve as critical indicators of pain and emotional states in horses. To overcome this limitation, the work proposes a video-based automated detection approach and presents the first systematic comparison of three methods: YOLOv12, optical flow magnitude thresholding, and a fine-tuned VideoMAE model for blink recognition in horses. Evaluated on a public dataset, the proposed framework achieves a binary classification accuracy of 0.926 and a macro F1-score of 0.898, demonstrating the feasibility of fine-grained micro-expression recognition in animals. These results advance the application of facial action unit analysis in animal welfare monitoring and provide a foundation for non-invasive, real-time assessment of equine affective states.

Abstract

Automated detection of equine facial action units (AUs) is a promising yet under-explored avenue for pain and affective state assessment in horses. Half and full-blink movements are recognised indicators of pain and stress, but as micro-expressions, their subtle, fine-grained nature makes them easily missed by the naked eye and only discernible through frame-by-frame video inspection, making reliable automated detection from video a particularly demanding task. We develop and evaluate three methods for automated blink classification from horse videos: a frame-based YOLOv12 detector, an optical flow magnitude thresholding approach, and a fine-tuned VideoMAE model, tested on a publicly available dataset. We achieve a macro-F1 score of 0.898 when doing blink classification and 0.926 on binary blink detection. Our results highlight both the potential and the inherent challenges of fine-grained AU detection for equine welfare monitoring.

Problem

• horse eye blink
• facial action units
• affective state assessment
• micro-expressions
• automated detection

Innovation

• blink detection
• equine facial action units
• VideoMAE
• YOLOv12
• optical flow