Point Transformer
This work addresses the challenge of handling unstructured 3D data for computer vision applications, though it appears incremental as it builds on existing transformer and point cloud methods.
The authors tackled the problem of processing unordered point sets by introducing Point Transformer, a deep neural network that uses local-global attention and SortNet to achieve permutation invariance, resulting in competitive performance on classification and part segmentation benchmarks.
In this work, we present Point Transformer, a deep neural network that operates directly on unordered and unstructured point sets. We design Point Transformer to extract local and global features and relate both representations by introducing the local-global attention mechanism, which aims to capture spatial point relations and shape information. For that purpose, we propose SortNet, as part of the Point Transformer, which induces input permutation invariance by selecting points based on a learned score. The output of Point Transformer is a sorted and permutation invariant feature list that can directly be incorporated into common computer vision applications. We evaluate our approach on standard classification and part segmentation benchmarks to demonstrate competitive results compared to the prior work. Code is publicly available at: https://github.com/engelnico/point-transformer