Active 3-D Shape Cosegmentation With Graph Convolutional Networks (original) (raw)

We present a novel active learning approach for shape cosegmentation based on graph convolutional networks (GCNs). The premise of our approach is to represent the collections of three-dimensional shapes as graph-structured data, where each node in the graph corresponds to a primitive patch of an oversegmented shape, and is associated with a representation initialized by extracting features. Then, the GCN operates directly on the graph to update the representation of each node based on a layer-wise propagation rule, which aggregates information from its neighbors, and predicts the labels for unlabeled nodes. Additionally, we further suggest an active learning strategy that queries the most informative samples to extend the initial training samples of GCN to generate more accurate predictions of our method. Our experimental results on the Shape COSEG dataset demonstrate the effectiveness of our approach. & SEGMENTATION OF THREE-DIMENSIONAL (3-D) shapes is a fundamental operation in geometric modeling and shape analysis. Recently, researchers have observed that by segmenting a set of 3-D shapes as a whole into consistent parts one can infer more knowledge than from an individual shape. This is the problem of cosegmentation. The results of shape cosegmentation can be applied to various applications in computer graphics and vision, such as modeling, shape retrieval, texturing, etc.