MotionAGFormer: Enhancing 3D Human Pose Estimation with a Transformer-GCNFormer Network | ResearchPod