An Unsupervised Monocular Visual Odometry Based on Multi-Scale Modeling.

An Unsupervised Monocular Visual Odometry Based on Multi-Scale Modeling.
复制标题

基于多尺度建模的无监督单目视觉里程计。

DOI:
10.3390/s22145193
复制
发表时间:
2022-07-11
期刊:
Sensors (Basel, Switzerland)
影响因子:
--
通讯作者:
--
中科院分区:
其他
文献类型:
--
作者:

文献摘要

参考文献

相似文献

无监督深度学习方法在从单目视频中联合估计相机姿势和深度方面取得了巨大成功。然而,以前的方法大多忽略了多尺度信息的重要性,这是至关重要的姿态估计和深度估计,特别是当运动模式发生变化。本文提出了一个无监督的框架,单目视觉里程计(VO),可以模拟多尺度信息。该方法利用密集连接的无环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机制来有效地建模长程依赖性。这两种方法都可以对图像中不同尺度的物体进行建模,从而提高了VO的精度,特别是在旋转场景中。在KITTI数据集上进行的大量实验表明,我们的方法与其他最先进的基于无监督学习的单目方法相比具有竞争力,并且与监督或基于模型的方法相当。特别是,我们已经取得了国家的最先进的旋转估计的结果。
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.
DOI: 10.3390/s21238155
发表时间: 2021-12-06
期刊: Sensors (Basel, Switzerland)
影响因子: --
作者:
Gadipudi N;Elamvazuthi I;Lu CK;Paramasivam S;Su S
通讯作者: Su S
用于基于流的无监督视觉里程计的变压器引导几何模型
DOI: 10.1007/s00521-020-05545-8
发表时间: 2021-01-02
影响因子: 6
作者:
Li, Xiangyu;Hou, Yonghong;Li, Wanqing
通讯作者: Li, Wanqing
DOI: 10.1109/tpami.2017.2658577
发表时间: 2018-03-01
影响因子: 23.6
作者:
Engel, Jakob;Koltun, Vladlen;Cremers, Daniel
通讯作者: Cremers, Daniel
DOI: 10.1109/tro.2020.3001674
发表时间: 2020-12-01
影响因子: 7.8
作者:
Costante, Gabriele;Mancini, Michele
通讯作者: Mancini, Michele
DOI: 10.1007/s11263-021-01484-6
发表时间: 2021-06-18
影响因子: 19.5
作者:
Bian, Jia-Wang;Zhan, Huangying;Reid, Ian
通讯作者: Reid, Ian