Systems and methods for human body pose estimation

A system for use in body pose estimation, comprising:

• an input module configured to obtain training data that includes observation vector data and corresponding pose vector data for a plurality of images,
• wherein the observation vector data is representative of the images in observation space, and the pose vector data is representative of the same images in pose space;
• and
• one or more processor configured to compute, based on the training data, a model that includes parameters of mapping from the observation space to latent space, parameters of mapping from the latent space to the pose space, and parameters of the latent space;
• wherein the latent space has a lower dimensionality than the observation space, and
• wherein the latent space has a lower dimensionality than the pose space.

A computer program product, comprising a non-transitory computer-readable medium having computer program instructions embodied thereon to cause one or more processor to implement a method for use in body pose estimation, the method comprising:

• obtaining training data that includes observation vector data and corresponding pose vector data for a plurality of images,
• wherein the observation vector data is representative of the images in observation space, and the pose vector data is representative of the same images in pose space;
• and
• computing, based on the training data, a model that includes parameters of mapping from the observation space to latent space, parameters of mapping from the latent space to the pose space, and parameters of the latent space;
• wherein the latent space has a lower dimensionality than the observation space, and
• wherein the latent space has a lower dimensionality than the pose space.