360VO: Visual Odometry Using A Single 360 Camera

Huang, Huajian; Yeung, Sai-Kit

doi:10.1109/icra46639.2022.9812203

Cited by 14 publications

(8 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [10], a fisheye calibration model has been designed and used for multiple fisheye camera rigs, as shown in Fig. 17 In recent years, spherical images captured by dual-fisheye cameras are becoming more and more popular, which promotes the development of panoramic SLAM systems [105], [116], [119], [120]. [105] designed a versatile visual SLAM, termed OpenVSLAM, which implements perspective, fisheye, and spherical camera models and supports monocular, stereo, and RGB-D cameras, as shown in Fig.…”

Section: Image Orientationmentioning

confidence: 99%

See 1 more Smart Citation

3D reconstruction of spherical images: A review of techniques, applications, and prospects

Li¹,

Weng²,

You³

et al. 2023

Preprint

View full text Add to dashboard Cite

3D reconstruction plays an increasingly important role in modern photogrammetric systems. Conventional satellite or aerial-based remote sensing (RS) platforms can provide the necessary data sources for the 3D reconstruction of large-scale landforms and cities. Even with low-altitude UAVs (Unmanned Aerial Vehicles), 3D reconstruction in complicated situations, such as urban canyons and indoor scenes, is challenging due to frequent tracking failures between camera frames and high data collection costs. Recently, spherical images have been extensively used due to the capability of recording surrounding environments from one camera exposure. In contrast to perspective images with limited FOV (Field of View), spherical images can cover the whole scene with full horizontal and vertical FOV and facilitate camera tracking and data acquisition in these complex scenes. With the rapid evolution and extensive use of professional and consumergrade spherical cameras, spherical images show great potential for the 3D modeling of urban and indoor scenes. Classical 3D reconstruction pipelines, however, cannot be directly used for spherical images. Besides, there exist few software packages that are designed for the 3D reconstruction of spherical images. As a result, this research provides a thorough survey of the state-ofthe-art for 3D reconstruction of spherical images in terms of data acquisition, feature detection and matching, image orientation, and dense matching as well as presenting promising applications and discussing potential prospects. We anticipate that this study offers insightful clues to direct future research.

show abstract

Section: Image Orientationmentioning

confidence: 99%

“…[120] investigated varying feature detection and description techniques, including SPHORB and SSIFT, and compared their performance in spherical image orientation based on SLAM. Recently, [119] implemented a direct SLAM system that extends DSO with the spherical camera model to process equirectangular images. Table VII presents…”

Section: Image Orientationmentioning

confidence: 99%

3D reconstruction of spherical images: A review of techniques, applications, and prospects

Li¹,

Weng²,

You³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…For the second type of images, Sumikura et al [22] proposed OpenVSLAM, which is a VO framework supporting equirectangular images. Huang et al [23] proposed a novel direct visual odometry algorithm, called 360VO, which takes advantage of a single 360°camera for robust localization and mapping.…”

Section: B Large-fov Slammentioning

confidence: 99%

LF-VIO: A Visual-Inertial-Odometry Framework for Large Field-of-View Cameras with Negative Plane

Wang

Yang

Shi

et al. 2022

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

View full text Add to dashboard Cite

“…The resolution of the image is 6080 × 3040. Finally, we used a 360 SFM pipeline [Huang and Yeung 2022] to process 360 • images, while we only reserved keyframes extracted by SFM pipeline. In addition, we separated the keyframes into two sets, one set serves as training data and another serves as testing data with ratio 4:1.…”

Section: Datasetmentioning

confidence: 99%

360Roam: Real-Time Indoor Roaming Using Geometry-Aware ${360^\circ}$ Radiance Fields

Huang¹,

Chen²,

Zhang³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Recently neural rendering has attracted great attention and demonstrated impressive rendering quality. Such learned view synthesis methods exploit neural networks to implicitly represent the structure and appearance of captured objects. The neural radiance field, pioneered by NeRF [Mildenhall et al. 2020] is currently the most promising path. Instead of pursuing precise geometry reconstruction, NeRF utilizes volume rendering techniques and multi-layer perceptrons to regress density and view-dependent color per ray. By densely sampling along each ray, it can generate photorealistic results even when encountered with light reflection, transparent objects, and thin structures. To reduce the long training and inference time, the following NeRF-based methods Liu et al. 2020;Pumarola et al. 2021;] have made great efforts and modifications to increase rendering quality and speed. However, most of them focus on object-centric or small-scale scenes where placements of cameras are constrained during the

show abstract

360VO: Visual Odometry Using A Single 360 Camera

Cited by 14 publications

References 20 publications

3D reconstruction of spherical images: A review of techniques, applications, and prospects

3D reconstruction of spherical images: A review of techniques, applications, and prospects

LF-VIO: A Visual-Inertial-Odometry Framework for Large Field-of-View Cameras with Negative Plane

360Roam: Real-Time Indoor Roaming Using Geometry-Aware ${360^\circ}$ Radiance Fields

Contact Info

Product

Resources

About