You can see, in below, some sample images from the "JHU-Table Setting" dataset. This dataset currently has about 2500 fully annotated table setting images. They are annotated using the Labelme annotation toolbox. You may refer to the following paper for more information about this toolbox:

B. C. Russell, A. Torralba, K. P. Murphy, W. T. Freeman,
LabelMe: a database and web-based tool for image annotation.
International Journal of Computer Vision, pages 157-173, Volume 77, Numbers 1-3, May, 2008.

For more information about the "JHU-Table Setting" dataset please contact ehsanj@jhu.edu.