Dataset for IKEA 3D models and aligned images

Joseph J. Lim, Hamed Pirsiavash, and Antonio Torralba


If you use our dataset and/or code, please cite this paper:
  • "Parsing IKEA Objects: Fine Pose Estimation." Joseph J. Lim, Hamed Pirsiavash, and Antonio Torralba. ICCV 2013.

       title={{Parsing IKEA Objects: Fine Pose Estimation}},
       author={Joseph J. Lim and Hamed Pirsiavash and Antonio Torralba},
There is a recent work providing extra 3D keypoint labels:


In order to use our database, you need to download the dataset and the toolbox.


In order to develop and evaluate fine pose estimation based on 3D models, we created a new dataset of images and 3D models representing typical indoor scenes. We explicitly collected IKEA 3D models from Google 3D Warehouse, and images from Flickr.

Our dataset contains about 759 images and 219 3D models. All 759 images are annotated using available models (about 90 different models). Also, we separate the data into two different splits: IKEAobject and IKEAroom.

  • Queried by individual object name (e.g. 'ikea chair poang' and 'ikea sofa ektorp')
  • Rather a simple scene with a few objects at relatively large scales
  • 288 images
  • Queried by 'ikea room' and 'ikea home'
  • This split contains more complex scene where multiple objects appear at a smaller scale
  • 471 images

For alignment, we created an online tool that allows an user to browse through models and label point correspondences (usually 5 are sufficient), and check the model's estimated pose as the user labels. Given these correspondences, we solve the least square problem using Levenberg-Marquardt. Here, we obtain the full intrinsic/extrinsic parameters except the skewness and principal points.


This work is funded by ONR MURI N000141010933. We also thank Phillip Isola, Aditya Khosla, Andrew Owens, and Carl Vondrick for important suggestions and discussion.

Our toolbox includes the following code/toolboxes:

  1. Joseph J. Lim, C. Lawrence Zitnick, and Piotr Dollar. Sketch Tokens: A Learned Mid-level Representation for Contour and Object Detection. CVPR 2013.
  2. Piotr Dollar. Piotr's Image and Video Matlab Toolbox.
  3. V. Lepetit, F. Moreno-Noguer and P. Fua. EPnP: An Accurate O(n) Solution to the PnP Problem. IJCV 2009.


If you find any bug or have a question, please contact Joseph Lim (lim AT