“…For our experiments, we use the Pascal 3D+ dataset (release 1.1) [22], which has 3D pose annotations for 12 common categories of interest: aeroplane (aero), bicycle (bike), boat, bottle, bus, car, chair, diningtable (dtable), motorbike (mbike), sofa, train, and tvmonitor (tv). The annotations are available for both VOC 2012 [1] and ImageNet [6] images.…”