Crowdsourced 3D Mapping: A Combined Multi-View Geometry and Self-Supervised Learning Approach

Chawla, Hemang; Jukola, Matti; Brouns, Terence; Zonooz, Bahram

doi:10.1109/iros45743.2020.9341243

Cited by 6 publications

(6 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To further evaluate our proposed system, we empirically study the impact of the number of turns used, as well as the frame rate of the videos on the calibration performance. We show that our system is superior to (Santana-Cedrés et al, 2017) as well as (Chawla et al, 2020a;Chawla et al, 2020b), which in turn outperforms the self-supervised (Gor-don et al, 2019). Finally, we demonstrate the application of our system for chessboard-free accurate monocular dense depth and ego-motion estimation on uncalibrated videos.…”

Section: Methodsmentioning

confidence: 58%

“…However, their applicability is constrained by the variety of images with different combinations of ground truth camera parameters used in training. On the otherhand self-supervised methods (Gordon et al, 2019) do not achieve similar performance (Chawla et al, 2020b). SfM has also been utilized to estimate camera parameters from a crowdsourced collection of images (Schonberger and Frahm, 2016).…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Practical Auto-Calibration for Spatial Scene-Understanding from Crowdsourced Dashcamera Videos

Chawla,

Jukola,

Marzban

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Spatial scene-understanding, including dense depth and ego-motion estimation, is an important problem in computer vision for autonomous vehicles and advanced driver assistance systems. Thus, it is beneficial to design perception modules that can utilize crowdsourced videos collected from arbitrary vehicular onboard or dashboard cameras. However, the intrinsic parameters corresponding to such cameras are often unknown or change over time. Typical manual calibration approaches require objects such as a chessboard or additional scene-specific information. On the other hand, automatic camera calibration does not have such requirements. Yet, the automatic calibration of dashboard cameras is challenging as forward and planar navigation results in critical motion sequences with reconstruction ambiguities. Structure reconstruction of complete visualsequences that may contain tens of thousands of images is also computationally untenable. Here, we propose a system for practical monocular onboard camera auto-calibration from crowdsourced videos. We show the effectiveness of our proposed system on the KITTI raw, Oxford RobotCar, and the crowdsourced D 2 -City datasets in varying conditions. Finally, we demonstrate its application for accurate monocular dense depth and ego-motion estimation on uncalibrated videos.

show abstract

Section: Methodsmentioning

confidence: 58%

Section: Related Workmentioning

confidence: 99%

Practical Auto-Calibration for Spatial Scene-Understanding from Crowdsourced Dashcamera Videos

Chawla,

Jukola,

Marzban

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, the process of camera calibration typically demands expertise, relying on experimentation and calculations to acquire the camera's intrinsic and extrinsic parameters. Therefore, some researchers have proposed HD map-construction methods that do not rely on the intrinsic and extrinsic parameters of cameras or smartphones [36][37][38]. Chawla et al proposed a method for extracting 3D positions of landmarks in HD maps [36].…”

Section: Camera-based Data-collection Methodsmentioning

confidence: 99%

A Review of Crowdsourcing Update Methods for High-Definition Maps

Guo,

Zhou,

et al. 2024

IJGI

View full text Add to dashboard Cite

High-definition (HD) maps serve as crucial infrastructure for autonomous driving technology, facilitating vehicles in positioning, environmental perception, and motion planning without being affected by weather changes or sensor-visibility limitations. Maintaining precision and freshness in HD maps is paramount, as delayed or inaccurate information can significantly impact the safety of autonomous vehicles. Utilizing crowdsourced data for HD map updating is widely recognized as a superior method for preserving map accuracy and freshness. Although it has garnered considerable attention from researchers, there remains a lack of comprehensive exploration into the entire process of updating HD maps through crowdsourcing. For this reason, it is imperative to review and discuss crowdsourcing techniques. This paper aims to provide an overview of the overall process of crowdsourced updates, followed by a detailed examination and comparison of existing methodologies concerning the key techniques of data collection, information extraction, and change detection. Finally, this paper addresses the challenges encountered in crowdsourced updates for HD maps.

show abstract

“…Moreover, self-supervised monocular depth estimation still requires prior knowledge of the camera intrinsics (focal length and principal point) during training, which may be different for each data source, may change over time, or be unknown a priori (Chawla et al, 2020). While multiple approaches to supervised camera intrinsics estimation have been proposed (Lopez et al, 2019;Zhuang et al, 2019), not many self-supervised approaches exist (Gordon et al, 2019).…”

Section: Related Workmentioning

confidence: 99%

Transformers in Self-Supervised Monocular Depth Estimation with Unknown Camera Intrinsics

Arnav¹,

Chawla²,

Zonooz³

2022

Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Application

Self Cite

View full text Add to dashboard Cite

The advent of autonomous driving and advanced driver assistance systems necessitates continuous developments in computer vision for 3D scene understanding. Self-supervised monocular depth estimation, a method for pixel-wise distance estimation of objects from a single camera without the use of ground truth labels, is an important task in 3D scene understanding. However, existing methods for this task are limited to convolutional neural network (CNN) architectures. In contrast with CNNs that use localized linear operations and lose feature resolution across the layers, vision transformers process at constant resolution with a global receptive field at every stage. While recent works have compared transformers against their CNN counterparts for tasks such as image classification, no study exists that investigates the impact of using transformers for self-supervised monocular depth estimation. Here, we first demonstrate how to adapt vision transformers for self-supervised monocular depth estimation. Thereafter, we compare the transformer and CNN-based architectures for their performance on KITTI depth prediction benchmarks, as well as their robustness to natural corruptions and adversarial attacks, including when the camera intrinsics are unknown. Our study demonstrates how transformer-based architecture, though lower in run-time efficiency, achieves comparable performance while being more robust and generalizable.

show abstract

Crowdsourced 3D Mapping: A Combined Multi-View Geometry and Self-Supervised Learning Approach

Cited by 6 publications

References 43 publications

Practical Auto-Calibration for Spatial Scene-Understanding from Crowdsourced Dashcamera Videos

Practical Auto-Calibration for Spatial Scene-Understanding from Crowdsourced Dashcamera Videos

A Review of Crowdsourcing Update Methods for High-Definition Maps

Transformers in Self-Supervised Monocular Depth Estimation with Unknown Camera Intrinsics

Contact Info

Product

Resources

About