Learning Monocular Visual Odometry through Geometry-Aware Curriculum Learning

Saputra, Muhamad Risqi U.; Gusmão, Pedro P. B. de; Wang, Sen; Markham, Andrew; Trigoni, Niki

doi:10.1109/icra.2019.8793581

Cited by 49 publications

(40 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, they have a framework with a pooling layer that makes their method more robust to various image qualities. Risqi et al [55] includes geometrical loss restrictions in order to increase consistency between multiple poses. In addition, Xue et al [56] implemented a memory module for storing global information and a refinement module for enhancing pose estimation.…”

Section: B Data-driven Methodsmentioning

confidence: 99%

LSTM and Filter Based Comparison Analysis for Indoor Global Localization in UAVs

et al. 2021

View full text Add to dashboard Cite

Deep learning (DL) based localization and Simultaneous Localization and Mapping (SLAM) has recently gained considerable attention demonstrating remarkable results. Instead of constructing handcrafted algorithms through geometric theories, DL based solutions provide a data-driven solution to the problem. Taking advantage of large amounts of training data and computing capacity, these approaches are increasingly developing into a new field that offers accurate and robust localization systems. In this work, the problem of global localization for unmanned aerial vehicles (UAVs) is analyzed by proposing a sequential, end-to-end, and multimodal deep neural network based monocular visual-inertial localization framework. More specifically, the proposed neural network architecture is threefold ; a visual feature extractor convNet network, a small IMU integrator bi-directional long short-term memory (LSTM), and a global pose regressor bi-directional LSTM network for pose estimation. In addition, by fusing the traditional IMU filtering methods instead of LSTM with the convNet, a more time-efficient deep pose estimation framework is presented. It is worth pointing out that the focus in this study is to evaluate the precision and efficiency of visual-inertial (VI) based localization approaches concerning indoor scenarios. The proposed deep global localization is compared with the various state-of-the-art algorithms on indoor UAV datasets, simulation environments and real-world drone experiments in terms of accuracy and time-efficiency. In addition, the comparison of IMU-LSTM and IMU-Filter based pose estimators is also provided by a detailed analysis. Experimental results show that the proposed filter-based approach combined with a DL approach has promising performance in terms of accuracy and time efficiency in indoor localization of UAVs.

show abstract

Section: B Data-driven Methodsmentioning

confidence: 99%

LSTM and Filter Based Comparison Analysis for Indoor Global Localization in UAVs

et al. 2021

View full text Add to dashboard Cite

show abstract

“…• The experiment results provided a competitive pose prediction compared to state-of-the-art monocular geometric [134], [135] and learning methods [121], [136]. Thus, they encouraged researchers to further explore and investigate learning-based VO methods.…”

Section: Fb-vo Siftmentioning

confidence: 95%

State of the Art in Vision-Based Localization Techniques for Autonomous Navigation Systems

2021

View full text Add to dashboard Cite

Vision-based localization systems, namely visual odometry (VO) and visual inertial odometry (VIO), have attracted great attention recently. They are regarded as critical modules for building fully autonomous systems. The simplicity of visual and inertial state estimators, along with their applicability in resource-constrained platforms motivated robotic community to research and develop novel approaches that maximize their robustness and reliability. In this paper, we surveyed state-of-the-art VO and VIO approaches. In addition, studies related to localization in visually degraded environments are also reviewed. The reviewed VO techniques and related studies have been analyzed in terms of key design aspects including appearance, feature, and learning based approaches. On the other hand, research studies related to VIO have been categorized based on the degree and type of fusion process into loosely-coupled, semi-tightly coupled, or tightly-coupled approaches and filtering or optimization-based paradigms. This paper provides an overview of the main components of visual localization, key design aspects highlighting the pros and cons of each approach, and compares the latest research works in this field. Finally, a detailed discussion of the challenges associated with the reviewed approaches and future research considerations are formulated.

show abstract

“…Clark et al [ 26 ] proposed to use a CNN–recurrent neural network(RNN) model to regress the camera pose from the monocular image sequence. Muhamad et al [ 27 ] applied curriculum learning to the geometric problem of the monocular VO system and proposed a geometry-aware objective function to regress the six-DoF camera pose. Wang et al [ 28 ] proposed the DeepVO, which utilizes a combination of CNN and RNN to estimate directly poses from the raw RGB image.…”

Section: Related Workmentioning

confidence: 99%

Stereo Visual Odometry Pose Correction through Unsupervised Deep Learning

Zhang

et al. 2021

Sensors

View full text Add to dashboard Cite

Visual simultaneous localization and mapping (VSLAM) plays a vital role in the field of positioning and navigation. At the heart of VSLAM is visual odometry (VO), which uses continuous images to estimate the camera’s ego-motion. However, due to many assumptions of the classical VO system, robots can hardly operate in challenging environments. To solve this challenge, we combine the multiview geometry constraints of the classical stereo VO system with the robustness of deep learning to present an unsupervised pose correction network for the classical stereo VO system. The pose correction network regresses a pose correction that results in positioning error due to violation of modeling assumptions to make the classical stereo VO positioning more accurate. The pose correction network does not rely on the dataset with ground truth poses for training. The pose correction network also simultaneously generates a depth map and an explainability mask. Extensive experiments on the KITTI dataset show the pose correction network can significantly improve the positioning accuracy of the classical stereo VO system. Notably, the corrected classical stereo VO system’s average absolute trajectory error, average translational relative pose error, and average translational root-mean-square drift on a length of 100–800 m in the KITTI dataset is 13.77 cm, 0.038 m, and 1.08%, respectively. Therefore, the improved stereo VO system has almost reached the state of the art.

show abstract

Learning Monocular Visual Odometry through Geometry-Aware Curriculum Learning

Cited by 49 publications

References 21 publications

LSTM and Filter Based Comparison Analysis for Indoor Global Localization in UAVs

LSTM and Filter Based Comparison Analysis for Indoor Global Localization in UAVs

State of the Art in Vision-Based Localization Techniques for Autonomous Navigation Systems

Stereo Visual Odometry Pose Correction through Unsupervised Deep Learning

Contact Info

Product

Resources

About