# Outdoor Motion Capture at Scale

## TL;DR

A new outdoor motion capture system using automated detection improves large-scale biomechanical data collection for sports like skiing.

## Contribution

The novel pipeline automates reference point and skier-specific keypoint detection, enabling efficient large-volume outdoor motion capture.

## Key findings

- Automated reference point detection reduced 3D segment-length variation by 23% compared to manual digitization.
- The skier-specific keypoint model achieved 98% PCK and lowered 3D segment-length variation by 0.5 cm compared to manual methods.

## Abstract

What are the main findings?
An outdoor motion capture pipeline was developed for large capture volumes using pan–tilt–zoom cameras.Automated detection of reference points and skier-specific keypoints improved 3D reconstruction consistency compared with manual digitization.

An outdoor motion capture pipeline was developed for large capture volumes using pan–tilt–zoom cameras.

Automated detection of reference points and skier-specific keypoints improved 3D reconstruction consistency compared with manual digitization.

What are the implications of the main findings?
The pipeline reduces manual post-processing considerably and makes large-scale biomechanical data collection in outdoor sports feasible.The approach supports the creation of sport-specific datasets for biomechanics and future 3D human pose estimation models.

The pipeline reduces manual post-processing considerably and makes large-scale biomechanical data collection in outdoor sports feasible.

The approach supports the creation of sport-specific datasets for biomechanics and future 3D human pose estimation models.

Capturing kinematic data in outdoor sports is challenging, as motions span large capture volumes and occur under difficult environmental conditions. Video-based approaches, particularly with pan–tilt–zoom cameras, offer a practical solution, but the extensive manual post-processing required limits their use to short sequences and few athletes. This study presents a motion capture pipeline that automates the detection of both reference points and sport-specific keypoints to overcome this limitation. The field test employed eight cameras covering a 250×80×30 m capture volume with nearly 300 reference points. Ten state-certified ski instructors performed eight standardized maneuvers. Reference points were localized through a hybrid approach combining YOLO object detection and ArUco marker identification. AlphaPose was fine-tuned on a new manually annotated dataset to detect skier-specific keypoints (e.g., skis, poles) alongside anatomical landmarks. Continuous frame-wise calibration and 3D reconstruction were performed using Direct Linear Transformation. Evaluation compared automated detections with manual annotations. Automated reference point detection achieved a mean localization error of 4.1 pixels (0.1% of 4K width) and reduced 3D segment-length variation by 23%. The skier-specific keypoint model reached 98% PCK, mAP of 0.97, and an MPJPE of 10.3 pixels while lowering 3D segment-length variation by 0.5 cm compared to manual digitization and 0.6 cm relative to a pretrained model. Replacing manual digitization with automated detection improves accuracy and facilitates kinematic data collection in large outdoor fields with many athletes and trials. The approach also enables the creation of sport-specific datasets valuable for biomechanical research and training next-generation 3D pose estimation models.