百科.dev
全部条目在线工具AI 编程趋势榜开源项目技术资讯提交条目
登录
< 返回工具列表
S

SampleNet

> 编程语言
开源

可分点云采样(CVPR 2020口述)

392 stars0 点赞0 次浏览
访问官网GitHub

工具介绍

可分点云采样(CVPR 2020口述)

SampleNet: Differentiable Point Cloud Sampling

Created by Itai Lang, Asaf Manor, and Shai Avidan from Tel Aviv University.

Introduction

This work is based on our arXiv tech report. Please read it for more information. You are also welcome to watch the oral talk from CVPR 2020.

There is a growing number of tasks that work directly on point clouds. As the size of the point cloud grows, so do the computational demands of these tasks. A possible solution is to sample the point cloud first. Classic sampling approaches, such as farthest point sampling (FPS), do not consider the downstream task. A recent work showed that learning a task-specific sampling can improve results significantly. However, the proposed technique did not deal with the non-differentiability of the sampling operation and offered a workaround instead.

We introduce a novel differentiable relaxation for point cloud sampling. Our approach employs a soft projection operation that approximates sampled points as a mixture of points in the primary input cloud. The approximation is controlled by a temperature parameter and converges to regular sampling when the temperature goes to zero. During training, we use a projection loss that encourages the temperature to drop, thereby driving every sample point to be close to one of the input points.

This approximation scheme leads to consistently good results on various applications such as classification, retrieval, and geometric reconstruction. We also show that the proposed sampling network can be used as a front to a point cloud registration network. This is a challenging task since sampling must be consistent across two different point clouds. In all cases, our method works better than existing non-learned and learned sampling alternatives.

Citation

If you find our work useful in your research, please consider citing:

@InProceedings{lang2020samplenet,
  author = {Lang, Itai and Manor, Asaf and Avidan, Shai},
  title = {{SampleNet: Differentiable Point Cloud Sampling}},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  pages = {7578--7588},
  year = {2020}
}

Installation and usage

This project contains three sub-directories, each one is a stand-alone project with it's own instructions. Please see the README.md in classification, registration, and reconstruction directories.

Usage of SampleNet in another project

Classification and reconstruction projects are implemented in TensorFlow. Registration is implemented with PyTorch library. We provide here an example code snippet, for the usage of PyTorch implementation of SampleNet with any task. The source files of SampleNet for this example are in registration/src/ folder.

…

License

This project is licensed under the terms of the MIT license (see LICENSE for details).

Issues· 0 开放

查看全部 Issues在 GitHub 打开

暂无开放 Issues,或尚未同步最近议题。

> 标签

Pythoncvpr2020deep-learninggeometry-processingneural-network

暂无评论,来聊聊你的看法吧

> 工具信息

发布日期2026年8月1日
最后更新2026年9月17日
分类编程语言
定价开源

> 相关工具

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言