Thermal video enables respiration tracking in moving marmosets
Tracking Is Not Sensing: Artifact-Aware Thermal Video Respiration Estimation in Head-Free Marmosets
Computer Vision and Pattern Recognition
Summary
Measuring breathing in small, freely-moving marmosets helps scientists study behavior and brain activity without disturbing the animals. The authors found that tracking just the position of the marmoset’s nose in thermal videos isn’t enough to get accurate breathing data because movement and fur confuse the measurements. They developed a new method called STAR that helps focus on breathing signals in the video, ignoring motion artifacts. This method improved breathing rate estimates enormously compared to simpler tracking.
What this means in practice
- •For biomedical device developers: Develop noninvasive breathing monitors for small primates in naturalistic or behavioral research settings that avoid sensors attached to the animal.
- •For animal welfare technology providers: Improve automated physiological monitoring tools for lab animals by integrating thermal video analysis that works despite animal head movement.$Commercial implications: Enables development of commercial contact-free respiration monitors for lab animals increasing welfare and reducing handling stress.
Authors
Penghao Wang, Sen Yang, Fenghua Xie, He Wang, Xindong Song
Abstract
Continuous physiological monitoring is essential for linking behavior, internal state, and neural activity in neuroscience and translational research. In common marmosets, a key nonhuman primate model for social neuroscience, respiratory rate provides a basic readout of arousal, stress, and physiological regulation. Contact sensors can provide accurate measurements, but they add preparation burden and may interfere with head-free behavioral protocols. Thermal video offers a noncontact alternative by capturing breathing-related temperature changes around the nose and mouth. However, in head-free marmosets, these cues are weak, localized, and easily corrupted by head motion, pose changes, and fur. Here we show that this setting exposes a broader challenge in video-based physiological sensing: visually reliable tracking does not necessarily imply physiologically valid measurement. Using synchronized thermal video and contact respiration recordings, we show that marmoset respiration is recoverable from thermal video but strongly ROI-sensitive. Under head-free motion, visually plausible nasal tracking can still produce traces inconsistent with contact respiration, demonstrating that tracking is not sensing. We therefore estimate respiratory rate from learned temporal signals extracted from thermal clips and introduce Selective Temporal Attenuation Regularization (STAR). STAR selectively attenuates high-motion thermal discontinuities through local temporal imputation during training, regularizing models toward respiration-consistent dynamics while leaving inference unchanged on raw video. Across multiple deep-learning backbones, STAR improves head-free estimation, reaching 2.23 bpm MAE compared with 20.76 bpm for tracked ROI analysis.