大工至善|大学至真分享 http://blog.sciencenet.cn/u/lcj2212916

博文

[转载]【计算机科学】【2020.06】三维点云目标跟踪的深度学习

已有 932 次阅读 2021-3-31 17:09 |系统分类:科研笔记|文章来源:转载

图片

本文为西班牙加泰罗尼亚政治大学(作者:Jaume Colom Hernandez)的硕士论文,共40页。

 

计算机视觉在图像和视频领域的研究已经取得了很大的进展,而激光雷达传感器的技术进步为计算机视觉研究开辟了一个全新的领域,已经广泛应用于交通、农业、医疗等行业。

 

本文主要研究三维点云中的目标跟踪问题。将成对的点云观测数据输入神经网络,以估计观测数据之间的姿态和平移。然后利用卡尔曼滤波和神经网络对估计值和外部特征进行处理,提取时空冗余信息,改善估计结果。该系统已在KITTI数据集上进行了测试,并对不同类型的物体和路径进行了预分段观测。结果表明,神经网络估计的位姿跟踪精度很高,但将估计的位姿和平移与递归神经网络相结合时,跟踪效果最好。

 

Great progress has been achieved in computer vision tasks within image and video, however technological advances in LiDAR sensors have created a whole new area of computer vision research devoted to it. With applications in many industries, such as transportation, agriculture or healthcare. This thesis studies object tracking in 3D point clouds. Pairs of point cloud observations are feed to a neural network to estimate pose and translation between the observations. Then this estimations, together with external features, are processed with Kalman Filter and RNN to extract spatial-temporal redundancies and improve the results. The system has been tested in the KITTI dataset, with pre-segmented observations, on different types of objects and paths. The results show that the neural network estimated pose gives a very accurate tracking, but the best results are achieved when combining the estimated pose and translations with a recurrent neural network.

 

1.       引言

2. 技术回顾

3. 研究方法

4. 结果

5. 预算

6. 结论


更多精彩文章请关注公众号:205328s611i1aqxbbgxv19.jpg




https://wap.sciencenet.cn/blog-69686-1279529.html

上一篇:[转载]【无人机】【2019】无人机巡线路由算法研究
下一篇:[转载]【信息技术】【2003】高质量音频信号的分析与编码
收藏 IP: 112.31.16.*| 热度|

0

该博文允许注册用户评论 请点击登录 评论 (0 个评论)

数据加载中...

Archiver|手机版|科学网 ( 京ICP备07017567号-12 )

GMT+8, 2024-3-28 16:58

Powered by ScienceNet.cn

Copyright © 2007- 中国科学报社

返回顶部