Autonomous Rotating Cameras Boost 3D Wildlife MoCap Yield without Human Operators
Amaan Vally, Daniel Joska, Naoya Muramatsu, Paul Amayo, Amir Patel
AI summary
Problem
Fixed cameras in outdoor settings have a limited field of view, making it difficult to capture continuous 3D markerless motion data of fast-moving animals without costly human operators or complex drone rigs.
Approach
A lightweight object detector tracks subjects at 10 Hz, which an Extended Kalman Filter bridges to a 50 Hz controller that autonomously rotates a camera to keep the target centered in the frame.
Key results
- +52% usable frames for running humans and +135% for free-running cheetahs
- Subject distribution shifted to image center, lowering theoretical 2D keypoint error
- Plug-and-play compatibility with existing 2D-to-3D reconstruction pipelines
- Robust tracking maintained during detector dropouts via EKF pseudo-observations
Why it matters
Provides a scalable, low-cost solution for collecting high-quality 3D animal kinematics in the wild, advancing biomechanics and biomimetic robotics research.
Abstract
We present a low-cost, autonomous, rotating- camera system that increases the usable data yield for 3D markerless motion capture of animals in uncontrolled outdoor settings. A lightweight detector (YOLOv4-Tiny) locates the subject at 10 Hz; an Extended Kalman Filter (EKF) bridges sparse detections to a 50 Hz full-state feedback (FSF) controller, keeping the subject centered without a human operator. The 3D reconstruction backend uses existing markerless 2D key- points and Full Trajectory Estimation (FTE) with a simple rotation compensation for moving cameras. On field videos of a running human and free-running cheetahs, the rotating cameras captured substantially more usable frames than fixed cameras: +52% for the human sequence (6593 vs. 4333 frames) and +135% across cheetah sequences (2419 vs. 1031 frames). Centering also shifts the subject-pixel distribution toward the image center, which theoretically lowers 2D keypoint error and thus 3D reprojection error for any pose-estimation backend. We detail the EKF design for sparse/noisy detections, the FSF controller with an integral state, and practical deployment considerations. Results show autonomous centering is a simple, deployable lever to scale outdoor animal motion capture without changing downstream reconstruction methods. markerless motion capture, rotating cameras, subject track- ing