Compositional Prototype Network with Multi-view Comparision for Few-Shot Point Cloud Semantic Segmentation
This work provides a method for few-shot point cloud semantic segmentation, which is crucial for reducing annotation costs in 3D vision applications, particularly benefiting researchers and practitioners dealing with scarce 3D data.
This paper addresses the problem of point cloud semantic segmentation with limited labeled data by proposing a Compositional Prototype Network. The network decomposes point cloud representations into local regional representations to calculate compositional prototypes and includes a Multi-View Comparison Component to exploit redundant views. The method significantly outperforms baselines on a new ScanNet-6i benchmark and boosts performance for few-shot classes in long-tail scenarios.
Point cloud segmentation is a fundamental visual understanding task in 3D vision. A fully supervised point cloud segmentation network often requires a large amount of data with point-wise annotations, which is expensive to obtain. In this work, we present the Compositional Prototype Network that can undertake point cloud segmentation with only a few labeled training data. Inspired by the few-shot learning literature in images, our network directly transfers label information from the limited training data to unlabeled test data for prediction. The network decomposes the representations of complex point cloud data into a set of local regional representations and utilizes them to calculate the compositional prototypes of a visual concept. Our network includes a key Multi-View Comparison Component that exploits the redundant views of the support set. To evaluate the proposed method, we create a new segmentation benchmark dataset, ScanNet-$6^i$, which is built upon ScanNet dataset. Extensive experiments show that our method outperforms baselines with a significant advantage. Moreover, when we use our network to handle the long-tail problem in a fully supervised point cloud segmentation dataset, it can also effectively boost the performance of the few-shot classes.