RoboPCA: Pose-centered Affordance Learning from Human Demonstrations for Robot Manipulation | Signal Canvas | ScienceToStartup