Elevation Angle Estimation in 2D Acoustic Images Using Pseudo Front View

Wang, Yusheng; Ji, Yonghoon; Liu, Dingyu; Yamashita, Atsushi; Asama, Hajime

doi:10.1109/lra.2021.3058911

Cited by 18 publications

(11 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Wang et al proposed a method for estimating the front depth of an acoustic image. It can solve the non-bijective correspondence problem [9]. However, the network was trained in a supervised manner, where ground truth labels were required.…”

Section: A Acoustic Camera 3d Reconstructionmentioning

confidence: 99%

“…Such methods only work under ideal conditions and are not general and robust. Recent deep learning-based methods have shown promising results for the single-view missing dimension estimation problem [8], [9]. The network in [9] was trained in a supervised manner, which required 3D supervision.…”

Section: Introductionmentioning

confidence: 99%

“…Recent deep learning-based methods have shown promising results for the single-view missing dimension estimation problem [8], [9]. The network in [9] was trained in a supervised manner, which required 3D supervision. Because acquiring groundtruth information for underwater scenes can be difficult due to the limitations of sensing techniques, a self-supervised training method would be highly valuable.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Learning Pseudo Front Depth for 2D Forward-Looking Sonar-based Multi-view Stereo

Wang

Asama

et al. 2022

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Self Cite

View full text Add to dashboard Cite

2D forward-looking sonar is a crucial sensor for underwater robotic perception. A well-known problem in this field is estimating missing information in the elevation direction during sonar imaging. There are demands to estimate 3D information per image for 3D mapping and robot navigation during fly-through missions. Recent learning-based methods have demonstrated their strengths, but there are still drawbacks. Supervised learning methods have achieved highquality results but may require further efforts to acquire 3D ground-truth labels. The existing self-supervised method requires pretraining using synthetic images with 3D supervision. This study aims to realize stable self-supervised learning of elevation angle estimation without pretraining using synthetic images. Failures during self-supervised learning may be caused by motion degeneracy problems. We first analyze the motion field of 2D forward-looking sonar, which is related to the main supervision signal. We utilize a modern learning framework and prove that if the training dataset is built with effective motions, the network can be trained in a self-supervised manner without the knowledge of synthetic data. Both simulation and real experiments validate the proposed method.

show abstract

Section: A Acoustic Camera 3d Reconstructionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Learning Pseudo Front Depth for 2D Forward-Looking Sonar-based Multi-view Stereo

Wang

Asama

et al. 2022

2022 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

Self Cite

View full text Add to dashboard Cite

show abstract

“…DeBortoli et al [30] proposed a self-supervised training procedure to fine-tune a Convolutional Neural Network (CNN) trained on simulated data with ground truth elevation information. Wang et al [31] use deep networks to transfer the acoustic view to a pseudo frontal view which was shown to help with estimating the elevation angle. However, these methods are limited to simple geometries or require collecting a larger dataset of real elevation data.…”

Section: A 3d Reconstruction Using Imaging Sonarmentioning

confidence: 99%

Neural Implicit Surface Reconstruction using Imaging Sonar

Qadri¹,

Kaess²,

Gkioulekas³

2022

Preprint

View full text Add to dashboard Cite

We present a technique for dense 3D reconstruction of objects using an imaging sonar, also known as forward-looking sonar (FLS). Compared to previous methods that model the scene geometry as point clouds or volumetric grids, we represent the geometry as a neural implicit function. Additionally, given such a representation, we use a differentiable volumetric renderer that models the propagation of acoustic waves to synthesize imaging sonar measurements. We perform experiments on real and synthetic datasets and show that our algorithm reconstructs high-fidelity surface geometry from multi-view FLS images at much higher quality than was possible with previous techniques and without suffering from their associated memory overhead.

show abstract

“…The drawback of those methods is that the reconstruction is sparse, as feature points are often difficult to extract in sonar images and correspondences can be reliably found only at nearby viewpoints. Generative models utilize the measured intensities and a known starting position in order to derive the slopes of the corresponding surfaces in the scene [3], [6], [8], [19], [22], [24]. While allowing a locally dense 3D reconstruction, these methods rely on the estimate of object edges and knowledge of the reflective 1 German Research Center for Artificial Intelligence, Bremen, Germany 2 Kraken Robotics, Bremen, Germany sarnold@ieee.org, bilal.wehbe@dfki.de properties of the surfaces.…”

Section: Introductionmentioning

confidence: 99%

Spatial Acoustic Projection for 3D Imaging Sonar Reconstruction

Arnold¹,

Wehbe²

2022

Preprint

View full text Add to dashboard Cite

In this work we present a novel method for reconstructing 3D surfaces using a multi-beam imaging sonar. We integrate the intensities measured by the sonar from different viewpoints for fixed cell positions in a 3D grid. For each cell we integrate a feature vector that holds the mean intensity for a discretized range of viewpoints. Based on the feature vectors and independent sparse range measurements that act as ground truth information, we train convolutional neural networks that allow us to predict the signed distance and direction to the nearest surface for each cell. The predicted signed distances can be projected into a truncated signed distance field (TSDF) along the predicted directions. Utilizing the marching cubes algorithm, a polygon mesh can be rendered from the TSDF. Our method allows a dense 3D reconstruction from a limited set of viewpoints and was evaluated on three real-world datasets.

show abstract

Elevation Angle Estimation in 2D Acoustic Images Using Pseudo Front View

Cited by 18 publications

References 21 publications

Learning Pseudo Front Depth for 2D Forward-Looking Sonar-based Multi-view Stereo

Learning Pseudo Front Depth for 2D Forward-Looking Sonar-based Multi-view Stereo

Neural Implicit Surface Reconstruction using Imaging Sonar

Spatial Acoustic Projection for 3D Imaging Sonar Reconstruction

Contact Info

Product

Resources

About