Hybrid 3D Human Pose Estimation with Monocular Video and Sparse IMUs
This work addresses occlusion and accuracy issues in human-centered computer vision, though it is incremental as it builds on existing sensor fusion approaches.
The paper tackled the challenge of integrating monocular video and sparse IMUs for 3D human pose estimation by proposing the RTOF framework, which reduced pose estimation error on the Total Capture dataset compared to baseline methods.
Temporal 3D human pose estimation from monocular videos is a challenging task in human-centered computer vision due to the depth ambiguity of 2D-to-3D lifting. To improve accuracy and address occlusion issues, inertial sensor has been introduced to provide complementary source of information. However, it remains challenging to integrate heterogeneous sensor data for producing physically rational 3D human poses. In this paper, we propose a novel framework, Real-time Optimization and Fusion (RTOF), to address this issue. We first incorporate sparse inertial orientations into a parametric human skeleton to refine 3D poses in kinematics. The poses are then optimized by energy functions built on both visual and inertial observations to reduce the temporal jitters. Our framework outputs smooth and biomechanically plausible human motion. Comprehensive experiments with ablation studies demonstrate its rationality and efficiency. On Total Capture dataset, the pose estimation error is significantly decreased compared to the baseline method.