Upright Orientation of Man-Made Objects

ACM Transaction on Graphics (Proceedings of SIGGRAPH 2008)

Hongbo Fu1    Daniel Cohen-Or2    Gideon Dror3    Alla Sheffer1

1University of British Columbia
Tel Aviv University
3Academic College of Tel-Aviv-Yaffo

Left: A man-made model with unnatural orientation. Middle: Six orientations obtained by aligning the model into a canonical coordinate frame using Principal Component Analysis. Right: Our method automatically detects the upright orientation of the model from its geometry alone.

Humans usually associate an upright orientation with objects, placing them in a way that they are most commonly seen in our surroundings. While it is an open challenge to recover the functionality of a shape from its geometry alone, this paper shows that it is often possible to infer its upright orientation by analyzing its geometry. Our key idea is to reduce the two-dimensional (spherical) orientation space to a small set of orientation candidates using functionality-related geometric properties of the object, and then determine the best orientation using an assessment function of several functional geometric attributes defined with respect to each candidate. Specifically we focus on obtaining the upright orientation for man-made objects that typically stand on some flat surface (ground, floor, table, etc.), which include the vast majority of objects in our everyday surroundings. For these types of models orientation candidates can be defined according to static equilibrium. For each candidate, we introduce a set of discriminative attributes linking shape to function. We learn an assessment function of these attributes from a training set using a combination of Random Forest classifier and Support Vector Machine classifier. Experiments demonstrate that our method generalizes well and achieves about 90% prediction accuracy for both a 10-fold cross-validation over the training set and a validation with an independent test set.

    PDF (1.9M)

    Divx Video (82.7M)

Examples of correct upright orientations found by our method. Click here to download the thumbnail images (12.2M) for the whole training and test sets with the failed models highlighted. The traning and test models are available upon request.
    author = {Hongbo Fu and Daniel Cohen-Or and Gideon Dror and Alla Sheffer},
    title = {Upright orientation of man-made objects},
    journal = {ACM Trans. Graph.},
    year = {2008},
    volume = {27},
    number = {3},

This research is supported in part by

  • National ICT Australia
  • The Israeli Ministry of Science
  • The Israel Science Foundation