DeepNet-Based 3D Visual Servoing Robotic Manipulation

Figure 5

Single input image and 2D key points: (a) nine vertices of the final stage and (b) output image with a bounding box.