milliFlow数据集
收藏资源简介:
由爱丁堡大学、麻省理工学院和伦敦大学学院联合开发的milliFlow数据集,旨在通过深度学习方法估计毫米波雷达点云中的场景流,以增强人体运动感知。该数据集以其隐私友好性和对智能家居应用的适用性而备受关注。它包含大规模多模态人体运动感知数据,涵盖了多种活动和不同环境,为研究者提供了丰富的实验材料。数据集的创建过程涉及商业Vayyar vTrigB成像毫米波雷达和RealSense D455深度相机的同步数据采集,确保了数据的多样性和准确性。milliFlow数据集不仅推动了毫米波雷达在人体运动感知领域的研究,也为智能系统决策、用户交互和个性化服务提供了强有力的支持。
The milliFlow dataset, jointly developed by the University of Edinburgh, MIT, and University College London, aims to estimate scene flow in millimeter-wave radar point clouds through deep learning methods to enhance human motion perception. This dataset has garnered attention for its privacy-friendly nature and suitability for smart home applications. It encompasses large-scale multimodal human motion perception data, covering a variety of activities and different environments, providing researchers with a wealth of experimental materials. The dataset creation process involves synchronized data acquisition using the commercial Vayyar vTrigB imaging millimeter-wave radar and the RealSense D455 depth camera, ensuring data diversity and accuracy. The milliFlow dataset not only advances research in the field of human motion perception using millimeter-wave radar but also provides robust support for intelligent system decision-making, user interaction, and personalized services.
milliFlow 数据集概述
数据集基本信息
- 名称: milliFlow
- 用途: 毫米波雷达点云上的场景流估计,用于人体运动感知
- 核心功能: 在传统毫米波雷达人体运动感知流程中提供点级运动信息
技术背景
- 技术基础: 基于毫米波雷达点云的场景流估计
- 应用领域: 人体运动感知
- 相关论文:
- 标题: "milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensing"
- 作者: Fangqiang Ding, Zhen Luo, Peijun Zhao, Chris Xiaoxuan Lu
- 会议: ECCV 2024
- arXiv链接: https://arxiv.org/pdf/2306.17010
数据集状态
- 最新动态:
- 2024-03-15: 预印本论文发布于arXiv
- 2024-07-01: 论文被ECCV 2024接收
- 2024-09-12: 演示视频和海报上线
使用信息
- 许可证:
- 代码和模型: CC BY-NC 4.0
- 数据集: CC BY-NC 4.0
- 引用格式: shell @InProceedings{Ding_2024_ECCV, author = {Ding, Fangqiang and Luo, Zhen and Zhao, Peijun and Lu, Chris Xiaoxuan}, title = {milliFlow: Scene Flow Estimation on mmWave Radar Point Cloud for Human Motion Sensing}, booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)}, year = {2024}, pages = {1-14} }
资源链接
- 论文PDF: https://arxiv.org/pdf/2306.17010
- 演示视频: https://youtu.be/fa91EeueGHA
- 使用指南: ./src/GETTING_STARTED.md




