fit_with_known_shape#

smplfitter.np.BodyFitter.fit_with_known_shape(shape_betas, target_vertices, target_joints=None, vertex_weights=None, joint_weights=None, kid_factor=None, num_iter=1, final_adjust_rots=True, initial_pose_rotvecs=None, scale_fit=False, requested_keys=('pose_rotvecs',))[source]#

Fits the body pose and translation with known shape parameters.

Parameters:

  • shape_betas – Shape parameters (betas), shaped as (batch_size, num_betas).

  • target_vertices – Target mesh vertices, shaped as (batch_size, num_vertices, 3).

  • target_joints – Optional target joint positions.

  • vertex_weights – Optional importance weights for vertices.

  • joint_weights – Optional importance weights for joints.

  • kid_factor – Optional kid blendshape factor.

  • num_iter – Number of fitting iterations.

  • final_adjust_rots – Whether to refine rotations after fitting.

  • initial_pose_rotvecs – Optional initial pose as rotation vectors.

  • scale_fit – Whether to estimate scale for fitted mesh.

  • requested_keys – List of result keys to return.

Returns:

Dictionary with pose_rotvecs, trans, and optionally other keys.