Learning Pose Grammar to Encode Human Body Configuration for 3D Pose Estimation
This addresses 3D pose estimation for computer vision applications, but it is incremental as it builds on existing methods with added constraints and simulation.
The paper tackles 3D human pose estimation by proposing a pose grammar model that learns a 2D-3D mapping with constraints like kinematics and symmetry, and it shows improved generalization in cross-view settings where other methods struggle.
In this paper, we propose a pose grammar to tackle the problem of 3D human pose estimation. Our model directly takes 2D pose as input and learns a generalized 2D-3D mapping function. The proposed model consists of a base network which efficiently captures pose-aligned features and a hierarchy of Bi-directional RNNs (BRNN) on the top to explicitly incorporate a set of knowledge regarding human body configuration (i.e., kinematics, symmetry, motor coordination). The proposed model thus enforces high-level constraints over human poses. In learning, we develop a pose sample simulator to augment training samples in virtual camera views, which further improves our model generalizability. We validate our method on public 3D human pose benchmarks and propose a new evaluation protocol working on cross-view setting to verify the generalization capability of different methods. We empirically observe that most state-of-the-art methods encounter difficulty under such setting while our method can well handle such challenges.