Zhi Henghui, Yin Chenyang, Li Huibin, Pang Shanmin
School of Mathematics and Statistics, Xi'an Jiaotong University, Xi'an 710049, China.
School of Software Engineering, Xi'an Jiaotong University, Xi'an 710049, China.
Sensors (Basel). 2022 Jul 11;22(14):5193. doi: 10.3390/s22145193.
Unsupervised deep learning methods have shown great success in jointly estimating camera pose and depth from monocular videos. However, previous methods mostly ignore the importance of multi-scale information, which is crucial for pose estimation and depth estimation, especially when the motion pattern is changed. This article proposes an unsupervised framework for monocular visual odometry (VO) that can model multi-scale information. The proposed method utilizes densely linked atrous convolutions to increase the receptive field size without losing image information, and adopts a non-local self-attention mechanism to effectively model the long-range dependency. Both of them can model objects of different scales in the image, thereby improving the accuracy of VO, especially in rotating scenes. Extensive experiments on the KITTI dataset have shown that our approach is competitive with other state-of-the-art unsupervised learning-based monocular methods and is comparable to supervised or model-based methods. In particular, we have achieved state-of-the-art results on rotation estimation.
无监督深度学习方法在从单目视频联合估计相机位姿和深度方面已取得巨大成功。然而，先前的方法大多忽略了多尺度信息的重要性，而多尺度信息对于位姿估计和深度估计至关重要，尤其是在运动模式发生变化时。本文提出了一种用于单目视觉里程计（VO）的无监督框架，该框架可以对多尺度信息进行建模。所提出的方法利用密集连接的空洞卷积来增加感受野大小而不丢失图像信息，并采用非局部自注意力机制来有效地对长距离依赖性进行建模。它们都可以对图像中不同尺度的物体进行建模，从而提高VO的准确性，尤其是在旋转场景中。在KITTI数据集上进行的大量实验表明，我们的方法与其他基于无监督学习的最先进单目方法具有竞争力，并且与基于监督或基于模型的方法相当。特别是，我们在旋转估计方面取得了最先进的结果。
