Monocular Depth Estimation in New Environments With Absolute Scale

Roussel, Tom; Eycken, Luc Van; Tuytelaars, Tinne

doi:10.1109/iros40897.2019.8967677

Cited by 16 publications

(13 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Methods addressing this problem add 3D-geometry-based losses to introduce scale-consistency [9], [10], yet utilize at least some depth or stereo supervision to introduce scaleawareness [11], [12]. Recently [26] introduced a similar instantaneous velocity based multi-modal supervision.…”

Section: Related Workmentioning

confidence: 99%

“…Consequently, most of the existing methods scale the estimated relative depth using the LiDAR ground truth during evaluation. Recent methods tackling this problem utilize additional 3D geometric constraints to introduce scale-consistency [9], [10], but require at least some depth or stereo supervision to predict at metric-scale [11], [12]. Nevertheless, obtaining metric scale predictions at low cost is necessary for practical deployment.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multimodal Scale Consistency and Awareness for Monocular Self-Supervised Depth Estimation

Chawla¹,

Arnav²,

Zonooz³

2021

2021 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

Dense depth estimation is essential to sceneunderstanding for autonomous driving. However, recent selfsupervised approaches on monocular videos suffer from scaleinconsistency across long sequences. Utilizing data from the ubiquitously copresent global positioning systems (GPS), we tackle this challenge by proposing a dynamically-weighted GPS-to-Scale (g2s) loss to complement the appearance-based losses. We emphasize that the GPS is needed only during the multimodal training, and not at inference. The relative distance between frames captured through the GPS provides a scale signal that is independent of the camera setup and scene distribution, resulting in richer learned feature representations. Through extensive evaluation on multiple datasets, we demonstrate scale-consistent and -aware depth estimation during inference, improving the performance even when training with low-frequency GPS data.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Multimodal Scale Consistency and Awareness for Monocular Self-Supervised Depth Estimation

Chawla¹,

Arnav²,

Zonooz³

2021

2021 IEEE International Conference on Robotics and Automation (ICRA)

View full text Add to dashboard Cite

show abstract

“…Although these methods achieve a great success, they cannot recover the metric depth and ego-motion. Some methods [38,43,36] also try to recover real scale of the scene by leveraging prior information such as height of the camera, etc.…”

Section: Structure From Motionmentioning

confidence: 99%

Road-aware Monocular Structure from Motion and Homography Estimation

Wei¹,

Chen²,

Zhang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Structure from motion (SFM) and ground plane homography estimation are critical to autonomous driving and other robotics applications. Recently, much progress has been made in using deep neural networks for SFM and homography estimation respectively. However, directly applying existing methods for ground plane homography estimation may fail because the road is often a small part of the scene. Besides, the performances of deep SFM approaches are still inferior to traditional methods. In this paper, we propose a method that learns to solve both problems in an end-to-end manner, improving performance on both. The proposed networks consist of a Depth-CNN, a Pose-CNN and a Ground-CNN. The Depth-CNN and Pose-CNN estimate dense depth map and ego-motion respectively, solving SFM, while the Pose-CNN and Ground-CNN followed by a homography layer solve the ground plane estimation problem. By enforcing coherency between SFM and homography estimation results, the whole network can be trained end to end using photometric loss and homography loss without any groundtruth except the road segmentation provided by an off-the-shelf segmenter. Comprehensive experiments are conducted on KITTI benchmark to demonstrate promising results compared with various state-of-the-art approaches.

show abstract

“…The ability of a robot to build a consistent map during its autonomous mission, widely known as Simultaneous Localization and Mapping (SLAM) [6], is strengthened when scale information is provided as robust visual odometry is generated [7]. Thus, depth-sensing is essential in any contemporary SLAM system [8], [9]. Commonly used sensors include LiDARs, binocular vision, etc., which are expensive and massive.…”

Section: Introductionmentioning

confidence: 99%

“…1. Various approaches [11], [22], [12], [18], [23] have been developed on the NYU Depth v2 dataset [24], which is an indoor environment without humans, while the KITTI vision suite collection [25] is selected for the outdoor cases [8], [22], [13]. Thus, no suitable data-sequence was available for our method's evaluation.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Monocular Human Depth Estimation and Segmentation on Embedded Systems

An¹,

Zhou²,

Yang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Estimating a scene's depth to achieve collision avoidance against moving pedestrians is a crucial and fundamental problem in the robotic field. This paper proposes a novel, low complexity network architecture for fast and accurate human depth estimation and segmentation in indoor environments, aiming to applications for resource-constrained platforms (including battery-powered aerial, micro-aerial, and ground vehicles) with a monocular camera being the primary perception module. Following the encoder-decoder structure, the proposed framework consists of two branches, one for depth prediction and another for semantic segmentation. Moreover, network structure optimization is employed to improve its forward inference speed. Exhaustive experiments on three selfgenerated datasets prove our pipeline's capability to execute in real-time, achieving higher frame rates than contemporary state-of-the-art frameworks (114.6 frames per second on an NVIDIA Jetson Nano GPU with TensorRT) while maintaining comparable accuracy.

show abstract

Monocular Depth Estimation in New Environments With Absolute Scale

Cited by 16 publications

References 19 publications

Multimodal Scale Consistency and Awareness for Monocular Self-Supervised Depth Estimation

Multimodal Scale Consistency and Awareness for Monocular Self-Supervised Depth Estimation

Road-aware Monocular Structure from Motion and Homography Estimation

Real-Time Monocular Human Depth Estimation and Segmentation on Embedded Systems

Contact Info

Product

Resources

About