Equine Pain Behavior Classification via Self-Supervised Disentangled Pose Representation
Equine Pain Behavior Classification via Self-Supervised Disentangled Pose Representation
复制标题
DOI：
10.1109/wacv51458.2022.00023
复制
发表时间：
2021-08
期刊：
影响因子：
--
通讯作者：
M. Rashid;S. Broomé;K. Ask;Elin Hernlund;P. Andersen;H. Kjellström;Yong Jae Lee
中科院分区：
文献类型：
--
作者：
M. Rashid;S. Broomé;K. Ask;Elin Hernlund;P. Andersen;H. Kjellström;Yong Jae Lee
及时发现马的疼痛对马的健康至关重要。马通过他们的面部和身体行为来表达疼痛，但可能会对不熟悉的人类观察者隐藏疼痛的迹象。此外，收集具有详细注释的马行为和疼痛状态的视觉数据既麻烦又不可扩展。因此，一个实用的马疼痛分类系统将使用未观察到的马的视频和弱标签。本文提出了一种马的疼痛分类方法，该方法使用未被观察到的马的多视图监控视频片段，这些视频片段具有暂时稀疏的视频级疼痛标签。为了确保疼痛仅从马的肢体语言中学习，我们首先训练一个自监督生成模型，在使用解纠缠的马姿势潜在表示进行疼痛分类之前，将其从外观和背景中解纠缠。为了充分利用疼痛标签，我们开发了一种新的损失方法，将疼痛分类作为一个多实例学习问题。该方法的疼痛分类准确率达到60%，优于人类专家。结果表明，习得的潜在马姿态表征是视点协变的，并且与马的外表分离。对疼痛分类片段的定性分析显示，我们的模型确定的疼痛症状与兽医实践中使用的马疼痛量表之间存在对应关系。
Timely detection of horse pain is important for equine welfare. Horses express pain through their facial and body behavior, but may hide signs of pain from unfamiliar human observers. In addition, collecting visual data with detailed annotation of horse behavior and pain state is both cumbersome and not scalable. Consequently, a pragmatic equine pain classification system would use video of the unobserved horse and weak labels. This paper proposes such a method for equine pain classification by using multi-view surveillance video footage of unobserved horses with induced orthopaedic pain, with temporally sparse video level pain labels. To ensure that pain is learned from horse body language alone, we first train a self-supervised generative model to disentangle horse pose from its appearance and background before using the disentangled horse pose latent representation for pain classification. To make best use of the pain labels, we develop a novel loss that formulates pain classification as a multi-instance learning problem. Our method achieves pain classification accuracy better than human expert performance with 60% accuracy. The learned latent horse pose representation is shown to be viewpoint covariant, and disentangled from horse appearance. Qualitative analysis of pain classified segments shows correspondence between the pain symptoms identified by our model, and equine pain scales used in veterinary practice.