Real-time 6D pose estimation from a single RGB image
Real-time 6D pose estimation from a single RGB image
复制标题
DOI：
10.1016/j.imavis.2019.06.013
复制
发表时间：
2019-09
期刊：
影响因子：
--
通讯作者：
Xin Zhang;Zhi-guo Jiang;Haopeng Zhang
中科院分区：
文献类型：
--
作者：
Xin Zhang;Zhi-guo Jiang;Haopeng Zhang
我们提出了一种端到端的深度学习架构，用于同时检测RGB图像中的对象并恢复6D姿态。具体来说，我们使用姿态估计模块扩展了2D检测管道，以根据2D检测结果间接回归对象3D顶点的图像坐标。然后，可以使用Perspective-n-Point算法来估计对象的6D姿态，而无需任何后细化。此外，我们精心设计了一个骨干结构，以保持空间分辨率的低级别特征的姿态估计任务。与最先进的基于RGB的姿态估计方法相比，我们的方法在两个基准数据集上以25 fps的推理速度在GTX 1080 Ti GPU上实现了具有竞争力或上级的性能，该GPU能够实时处理。
We propose an end-to-end deep learning architecture for simultaneously detecting objects and recovering 6D poses in an RGB image. Concretely, we extend the 2D detection pipeline with a pose estimation module to indirectly regress the image coordinates of the object's 3D vertices based on 2D detection results. Then the object's 6D pose can be estimated using a Perspective-n-Point algorithm without any post-refinements. Moreover, we elaborately design a backbone structure to maintain spatial resolution of low level features for pose estimation task. Compared with state-of-the-art RGB based pose estimation methods, our approach achieves competitive or superior performance on two benchmark datasets at an inference speed of 25 fps on a GTX 1080Ti GPU, which is capable of real-time processing.