Tracking the Flight: Exploring a Computational Framework for Analyzing Escape Responses in Plains Zebra (Equus quagga)
Ethological research increasingly benefits from the growing affordability and accessibility of drones, which enable the capture of high-resolution footage of animal movement at fine spatial and temporal scales. However, analyzing such footage presents the technical challenge of separating animal movement from drone motion.
Evidence notes
- The analyzed 44-zebra escape sequence was recorded by a DJI Mavic 3 Pro in 4K at 29.97 fps during field experiments in Mpala, Kenya.
- Flights were conducted at or above 60 m, mostly at 100 m, and the wider five-day collection produced 41 usable videos.
- Structure-from-Motion with linear interpolation produced the lowest trajectory-unwrapping error at 0.275 zebra body lengths, versus 0.910 for frame-to-frame registration.
Company context
Aerial imaging and UAS market leader; consumer and enterprise drone platforms, cameras, and flight control ecosystems.