Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

FANTrack: 3D Multi-Object Tracking with Feature Association Network #35

Open
DeepTecher opened this issue May 10, 2019 · 0 comments
Open
Labels
IEEE IV IEEE Intelligent Vehicles Symposium 多目标跟踪|MOT Multi-Object Tracking 论文速递 无人驾驶最新相关论文

Comments

@DeepTecher
Copy link
Owner

DeepTecher commented May 10, 2019

FANTrack: 3D Multi-Object Tracking with Feature Association Network
提交日期: 2019-05-07
研究机构:
作者:Erkan Baser, Venkateshwaran Balasubramanian, Prarthana Bhattacharyya, Krzysztof Czarnecki
摘要:我们提出了一种数据驱动的在线多目标跟踪(MOT)方法,该方法使用卷积神经网络(CNN)在逐个检测框架中进行数据关联。多目标跟踪的问题旨在将噪声检测分配给跨越一系列帧的先验未知和随时间变化的跟踪对象。大多数现有解决方案都侧重于繁琐地设计成本函数或将数据关联任务制定为可以有效解决的复杂优化问题。相反,我们利用深度学习的力量将数据关联问题表述为CNN中的推理。为此,我们建议学习一种相似性函数,该函数结合了来自对象的图像和空间特征的线索。我们的解决方案学会完全从数据中执行3D全局分配,处理嘈杂的检测和不同数量的目标,并且易于训练。我们在具有挑战性的KITTI数据集上评估我们的方法并显示出有竞争力的结果。代码:wise-lab/fantrack

@DeepTecher DeepTecher added 论文速递 无人驾驶最新相关论文 多目标跟踪|MOT Multi-Object Tracking IEEE IV IEEE Intelligent Vehicles Symposium labels May 10, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
IEEE IV IEEE Intelligent Vehicles Symposium 多目标跟踪|MOT Multi-Object Tracking 论文速递 无人驾驶最新相关论文
Projects
None yet
Development

No branches or pull requests

1 participant