Research outputs and technical references.
Public papers, project reports, dataset notes, and other technical evidence for our tracking systems.
arXiv preprint - 2026
Khan, Muhammad Saif Ullah and Wang, Chen-Yu and Prokosch, Tim and Lorenz, Micheal and Taetz, Bertram and Stricker, Didier
Introducing a fast and accurate graph-attention network for inverse kinematics that reconstructs full-body joint orientations from 3D joint positions.
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) - 2026
Khan, Muhammad Saif Ullah and Stricker, Didier
Introducing SIMSPINE, a biomechanics-aware simulation framework and dataset for 3D spine motion annotation, enabling data-driven learning of vertebral kinematics from natural full-body motions.
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops - 2025
Khan, Muhammad Saif Ullah and Krau{ss}, Stephan and Stricker, Didier
Introducing SpineTrack, a comprehensive dataset for 2D spine pose estimation in unconstrained settings, featuring detailed spinal keypoints and an active learning pipeline for anatomically consistent annotations at scale.