RPGD: 3D人間姿勢推定のためのRANSAC-P3P勾配降下法による外部キャリブレーション
RPGD: RANSAC-P3P Gradient Descent for Extrinsic Calibration in 3D Human Pose Estimation
自然な人間動作のみを用いて、モーションキャプチャの3D骨格データとRGBカメラを頑健に位置合わせする外部キャリブレーション手法を提案し、大規模データセットで高精度を実証した。
著者: Zhanyu Tuo
分類: cs.CV, cs.AI, cs.LG, cs.RO
原文アブストラクト
In this paper, we propose RPGD (RANSAC-P3P Gradient Descent), a human-pose-driven extrinsic calibration framework that robustly aligns MoCap-based 3D skeletal data with monocular or multi-view RGB cameras using only natural human motion. RPGD formulates extrinsic calibration as a coarse-to-fine problem tailored to human poses, combining the global robustness of RANSAC-P3P with Gradient-Descent-based refinement. We evaluate RPGD on three large-scale public 3D HPE datasets as well as on a self-collected in-the-wild dataset. Experimental results demonstrate that RPGD consistently recovers extrinsic parameters with accuracy comparable to the provided ground truth, achieving sub-pixel MPJPE reprojection error even in challenging, noisy settings. These results indicate that RPGD provides a practical and automatic solution for reliable extrinsic calibration of large-scale 3D HPE dataset collection.