Estimating the 6D object pose is one of the most fundamental problems in Computer Vision as it is essential for various applications, including robotic grasping and autonomous driving. Thereby, the 6D object pose describes the orientation and position of the object in 3D space and is often a key step-stone for further 3D reasoning and manipulation. Due to its high relevance for many applications, this dissertation focuses on the problem of 6D pose estimation from monocular data alone.