抄録
We introduce a new framework for the automatic selection of the best views of 3D models based on the assumption that models belonging to the same class of shapes share the same salient features. The main issue is learning these features. We propose an algorithm for computing these features and their corresponding saliency value. At the learning stage, a large set of features are computed from every model and a boosting algorithm is applied to learn the classification function in the feature space. AdaBoost learns a classifier that relies on a small subset of the features with the mean of weak classifiers, and provides an efficient way for feature selection and combination. Moreover it assigns weights to the selected features which we interpret as a measure of the feature saliency within the class. Our experiments using the LightField (LFD) descriptors and the Princeton Shape Benchmark show the suitability of the approach to 3D shape classification and best-view selection for online visual browsing of 3D data collections.
