Stratified Transformer for 3D Point Cloud Segmentation
By Xin Lai, Jianhui Liu, Li Jiang, Liwei Wang, Hengshuang Zhao, Shu Liu, Xiaojuan Qi, Jiaya Jia. Computer Vision and Pattern Recognition, 2022
400 citations recorded by the source corpus.
3D point cloud segmentation has made tremendous progress in recent years. Most current methods focus on aggregating local features, but fail to directly model long-range dependencies. In this paper, we propose Stratified Transformer that is able to capture long-range contexts and demonstrates strong generalization ability and high performance. Specifically, we first put forward a novel key sampling strategy. For each query point, we sample nearby points densely and distant points sparsely as its keys in a stratified way, which enables the model to enlarge the effective receptive field and enjoy long-range contexts at a low computational cost. Also, to combat the challenges posed by irregular point arrangements, we propose first-layer point embedding to aggregate local information, which facilitates convergence and boosts performance. Besides, we adopt contextual relative position encoding to adaptively capture position information. Finally, a memory-efficient implementation is introduced to overcome the issue of varying point numbers in each window. Extensive experiments demonstrate the effectiveness and superiority of our method on S3DIS, ScanNetv2 and ShapeNetPart datasets. Code is available at https://github.com/dvlab-research/Stratified-Transformer.
Subfields: Computational Mechanics, Environmental Engineering, Geology
View Paper: https://www.semanticscholar.org/paper/e259cfabd07b65d703bf8b768c667bb50796954d
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computational Mechanics
- Jiaya Jia
- Or Litany
- Shaoshuai Shi
- Bernard Ghanem
- Bastian Leibe
- Xiaojuan Qi
- Guocheng Qian
- Hasan Hammoud
In Environmental Engineering
- Shaoshuai Shi
- Jiaya Jia
- Dengxin Dai
- Bastian Leibe
- Yukang Chen
- Or Litany
- Bernt Schiele
- Jiangmiao Pang
In Geology
- Jiaya Jia
- Or Litany
- Shaoshuai Shi
- Francis Engelmann
- Yukang Chen
- Xiaojuan Qi
- Bastian Leibe
- Guocheng Qian
Citations in the corpus
- PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies (2022, methodology): “For PointNet++, we use the results reported from the Stratified Transformer [17] for comparison.”
- Transformer-Based Visual Segmentation: A Survey (2023, background): “Stratified Transformer [241] PCSS Point Cloud / Semantic Masks Pure transformer cube-wised point cloud transformer with mixed local and long-range key point sampling.”
- VoxelNeXt: Fully Sparse VoxelNet for 3D Object Detection and Tracking (2023, background): “, well-designed convolution [7], large kernels [8], and transformers [24, 25, 34].”
- Mask3D: Mask Transformer for 3D Semantic Instance Segmentation (2022, methodology): “domain though and recent Transformer-based methods focus on either 3D object detection [34, 37, 39] or 3D semantic segmentation [29, 40, 63].”
- Motion Transformer with Global Intention Localization and Local Movement Refinement (2022, methodology): “Our approach is inspired by DETR [4] and its follow-up works [63, 32, 56, 25, 28, 10, 59], especially DAB-DETR [28], where the object query is considered as the positional embedding of a spatial anchor box.”