https://arxiv.org/pdf/2403.14626.pdf
摘要:障碍物检测与跟踪是机器人自主导航的重要组成部分。
Abstract— Obstacle detection and tracking represent a critical component in robot autonomous navigation.
在本文中,我们提出了一种基于变压器的ODTFormer模型来解决障碍物检测和跟踪问题。
In this paper, we propose ODTFormer, a Transformer-based model to address both obstacle detection and tracking problems.
对于检测任务,我们的方法利用可变形的注意力来构建一个3D成本体,该成本体以体素占用网格的形式逐步解码。
For the detection task, our approach leverages deformable attention to construct a 3D cost volume, which is decoded progressively in the form of voxel occupancy grids.
我们通过匹配连续帧之间的体素进一步跟踪障碍物。
We further track the obstacles by matching the voxels between consecutive frames.
整个模型可以以端到端的方式进行优化。
The entire model can be optimized in an end-to-end manner.
通过在DrivingStereo和KITTI基准上的大量实验,我们的模型在障碍物检测任务中达到了最先进的性能。
Through extensive experiments on DrivingStereo and KITTI benchmarks, our model achieves state-of-the-art performance in the obstacle detection task.
我们还报告了与最先进的障碍物跟踪模型相当的准确性,而只需要其计算成本的一小部分,通常少10到20倍。
We also report comparable accuracy to state-of-the-art obstacle tracking models while requiring only a fraction of their computation cost, typically ten-fold to twentyfold less.
代码和模型权重将被公开发布。
The code and model weights will be publicly released. |