Collect-and-Distribute Transformer for 3D Point Cloud Analysis

• URL: http://arxiv.org/abs/2306.01257v2
• Date: Tue, 31 Oct 2023 03:55:12 GMT
• Title: Collect-and-Distribute Transformer for 3D Point Cloud Analysis
• Authors: Haibo Qiu, Baosheng Yu, Dacheng Tao
• Abstract summary: We propose a new transformer network equipped with a collect-and-distribute mechanism to communicate short- and long-range contexts of point clouds. Results show the effectiveness of the proposed CDFormer, delivering several new state-of-the-art performances on point cloud classification and segmentation tasks.
• Score: 82.03517861433849
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Remarkable advancements have been made recently in point cloud analysis through the exploration of transformer architecture, but it remains challenging to effectively learn local and global structures within point clouds. In this paper, we propose a new transformer network equipped with a collect-and-distribute mechanism to communicate short- and long-range contexts of point clouds, which we refer to as CDFormer. Specifically, we first employ self-attention to capture short-range interactions within each local patch, and the updated local features are then collected into a set of proxy reference points from which we can extract long-range contexts. Afterward, we distribute the learned long-range contexts back to local points via cross-attention. To address the position clues for short- and long-range contexts, we additionally introduce the context-aware position encoding to facilitate position-aware communications between points. We perform experiments on five popular point cloud datasets, namely ModelNet40, ScanObjectNN, ShapeNetPart, S3DIS and ScanNetV2, for classification and segmentation. Results show the effectiveness of the proposed CDFormer, delivering several new state-of-the-art performances on point cloud classification and segmentation tasks. The source code is available at \url{https://github.com/haibo-qiu/CDFormer}.

Related papers

• Few-Shot 3D Point Cloud Semantic Segmentation via Stratified Class-Specific Attention Based Transformer Network [22.9434434107516]
  We develop a new multi-layer transformer network for few-shot point cloud semantic segmentation. Our method achieves the new state-of-the-art performance, with 15% less inference time, over existing few-shot 3D point cloud segmentation models.
  arXiv  Detail & Related papers  (2023-03-28T00:27:54Z)
• Adaptive Edge-to-Edge Interaction Learning for Point Cloud Analysis [118.30840667784206]
  Key issue for point cloud data processing is extracting useful information from local regions. Previous works ignore the relation between edges in local regions, which encodes the local shape information. This paper proposes a novel Adaptive Edge-to-Edge Interaction Learning module.
  arXiv  Detail & Related papers  (2022-11-20T07:10:14Z)
• Graph Neural Network and Spatiotemporal Transformer Attention for 3D Video Object Detection from Point Clouds [94.21415132135951]
  We propose to detect 3D objects by exploiting temporal information in multiple frames. We implement our algorithm based on prevalent anchor-based and anchor-free detectors.
  arXiv  Detail & Related papers  (2022-07-26T05:16:28Z)
• SemAffiNet: Semantic-Affine Transformation for Point Cloud Segmentation [94.11915008006483]
  We propose SemAffiNet for point cloud semantic segmentation. We conduct extensive experiments on the ScanNetV2 and NYUv2 datasets.
  arXiv  Detail & Related papers  (2022-05-26T17:00:23Z)
• Stratified Transformer for 3D Point Cloud Segmentation [89.9698499437732]
  Stratified Transformer is able to capture long-range contexts and demonstrates strong generalization ability and high performance. To combat the challenges posed by irregular point arrangements, we propose first-layer point embedding to aggregate local information. Experiments demonstrate the effectiveness and superiority of our method on S3DIS, ScanNetv2 and ShapeNetPart datasets.
  arXiv  Detail & Related papers  (2022-03-28T05:35:16Z)
• Deep Point Cloud Reconstruction [74.694733918351]
  Point cloud obtained from 3D scanning is often sparse, noisy, and irregular. To cope with these issues, recent studies have been separately conducted to densify, denoise, and complete inaccurate point cloud. We propose a deep point cloud reconstruction network consisting of two stages: 1) a 3D sparse stacked-hourglass network as for the initial densification and denoising, 2) a refinement via transformers converting the discrete voxels into 3D points.
  arXiv  Detail & Related papers  (2021-11-23T07:53:28Z)
• Learning point embedding for 3D data processing [2.12121796606941]
  Current point-based methods are essentially spatial relationship processing networks. Our architecture, PE-Net, learns the representation of point clouds in high-dimensional space. Experiments show that PE-Net achieves the state-of-the-art performance in multiple challenging datasets.
  arXiv  Detail & Related papers  (2021-07-19T00:25:28Z)
• Compositional Prototype Network with Multi-view Comparision for Few-Shot Point Cloud Semantic Segmentation [47.0611707526858]
  A fully supervised point cloud segmentation network often requires a large amount of data with point-wise annotations. 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.
  arXiv  Detail & Related papers  (2020-12-28T15:01:34Z)
• Dense-Resolution Network for Point Cloud Classification and Segmentation [42.316932316581635]
  DRNet is designed to learn local point features from the point cloud in different resolutions. In addition to validating the network on widely used point cloud segmentation and classification benchmarks, we also test and visualize the performance of the components.
  arXiv  Detail & Related papers  (2020-05-14T06:13:53Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.