Studying the detailed biomechanics of flying animals requires accurate three-dimensional coordinates for key anatomical landmarks. Traditionally, this relies on manually digitizing animal videos, a labor-intensive task that scales poorly with increasing framerates and numbers of cameras. Here, we present a workflow that combines deep learning–powered automatic digitization with filtering and correction of mislabeled points using quality metrics from deep learning and 3D reconstruction. We tested our workflow using a particularly challenging scenario: bat flight. First, we documented four bats flying steadily in a 2 m³ wind tunnel test section. Wing kinematic parameters resulting from manually digitizing bats with markers applied to anatomical landmarks were not significantly different from those resulting from applying our workflow to the same bats without markers for five out of six parameters. Second, we compared coordinates from manual digitization against those yielded via our workflow for bats flying freely in a 344 m³ enclosure. Average distance between coordinates from our workflow and those from manual digitization was less than a millimeter larger than the average human-to-human coordinate distance. The improved efficiency of our workflow has the potential to increase the scalability of studies on animal flight biomechanics.

中文翻译：

---

基于深度学习的新型 3D 摄像工作流程在蝙蝠飞行中的应用


研究飞行动物的详细生物力学需要关键解剖标志的精确三维坐标。传统上，这依赖于手动数字化动物视频，这是一项劳动密集型任务，随着帧速率和摄像机数量的增加，扩展性很差。在这里，我们提出了一个工作流程，它将深度学习驱动的自动数字化与使用深度学习和 3D 重建的质量指标过滤和纠正错误标记点相结合。我们使用一个特别具有挑战性的场景来测试我们的工作流程：蝙蝠飞行。首先，我们记录了四只蝙蝠在 2 m ³ 风洞测试段中稳定飞行的情况。手动数字化带有应用于解剖标志的标记的蝙蝠所产生的机翼运动参数与将我们的工作流程应用于没有标记的相同蝙蝠所产生的结果没有显着差异，六分之五参数没有标记。其次，我们将手动数字化的坐标与通过工作流程生成的蝙蝠在 344 m ³ 外壳中自由飞行的坐标进行了比较。我们工作流程中的坐标与手动数字化中的坐标之间的平均距离比人与人之间的平均坐标距离大不到一毫米。我们工作流程效率的提高有可能提高动物飞行生物力学研究的可扩展性。