RGB-D Object Dataset The dataset contains 300 objects (aka instance) in 51 categories. The name of each object follows the format _, e.g. apple_1 is the first object (instance) in the "apple" category. Each instance consists of multiple RGB-D frames that come from three video sequences. This part of the dataset contains the ground truth pose labels for every image in the RGB-D Object Dataset, exactly as used in the pose recognition evaluation of our AAAI-11 paper (see below). NOTE: The pose labels are approximate. For each video sequence, the pose of the object in two images were annotated by hand and the pose of all other images in the sequence are labeled by interpolation, assuming constant rotation of the turntable. While this works well in most cases, the assumption of constant rotation is not always true for the turntable we used to collect the data. Windows users: The dataset was compressed into a tarball using Linux. Some Windows extractors have problems reading the files. One program that can be used to extract the data in Windows is 7zip ( http://www.7-zip.org/ ). Open the single file extracted from the tarball again with 7zip to unpack it into a directory. The files are named as follows: __