Event
RoPT: A Robust 6DoF Object Pose Tracking Framework for Videos
1 reports · 1 sources
Yongqing Liang et al. propose RoPT, a 6DoF object pose tracking framework for robot interaction that couples visual perception, temporal kinematics, and global relocalization into a closed loop to counter catastrophic divergence of monocular pipelines under fast motion. Its visual module tracks object masks to handle partial and severe occlusion; a kinematic factor graph regularizes inter-frame updates, predicts forward poses, and drives parallel multi-hypothesis verification to avoid local minima; an active relocalization module equips the system with failure awareness and deterministic self-recovery. The authors claim the framework achieves significantly higher tracking survival rates under fast motion on challenging benchmarks and in real robot demonstrations.
Report timeline
- RoPT: A Robust 6DoF Object Pose Tracking Framework for Video
RoPT is a 6DoF object pose tracking framework for robotic interaction that couples visual perception, temporal kinematics, and global re-localization into a closed loop to counter catastrophic divergence in monocular pipelines under fast motion. Its visual module tracks object masks to handle partial and severe occlusions; a kinematic factor graph regularizes frame-to-frame updates, predicts forward poses, and drives parallel multi-hypothesis verification to avoid local minima; and an active re-localization module gives the system failure awareness and deterministic self-recovery.
arXiv cs.RO ·