On the Uncertainty of Self-Supervised Monocular Depth Estimation

Poggi, Matteo; Aleotti, Filippo; Tosi, Fabio; Mattoccia, Stefano

doi:10.1109/cvpr42600.2020.00329

Cited by 210 publications

(112 citation statements)

References 54 publications

Supporting

Mentioning

110

Contrasting

Unclassified

Order By: Relevance

“…MC dropout has been widely adopted in various research fields. Applications include camera pose estimation (Kendall and Cipolla, 2016), depth estimation (Poggi et al, 2020), pedestrian localization (Bertoni et al, 2019), semantic segmentation (Mukhoti and Gal, 2018), and electrocardiogram signal detection (Elola et al, 2019).…”

Section: Related Workmentioning

confidence: 99%

A Bayesian Deep Neural Network for Safe Visual Servoing in Human–Robot Interaction

2021

View full text Add to dashboard Cite

Safety is an important issue in human–robot interaction (HRI) applications. Various research works have focused on different levels of safety in HRI. If a human/obstacle is detected, a repulsive action can be taken to avoid the collision. Common repulsive actions include distance methods, potential field methods, and safety field methods. Approaches based on machine learning are less explored regarding the selection of the repulsive action. Few research works focus on the uncertainty of the data-based approaches and consider the efficiency of the executing task during collision avoidance. In this study, we describe a system that can avoid collision with human hands while the robot is executing an image-based visual servoing (IBVS) task. We use Monte Carlo dropout (MC dropout) to transform a deep neural network (DNN) to a Bayesian DNN, and learn the repulsive position for hand avoidance. The Bayesian DNN allows IBVS to converge faster than the opposite repulsive pose. Furthermore, it allows the robot to avoid undesired poses that the DNN cannot avoid. The experimental results show that Bayesian DNN has adequate accuracy and can generalize well on unseen data. The predictive interval coverage probability (PICP) of the predictions along x, y, and z directions are 0.84, 0.94, and 0.95, respectively. In the space which is unseen in the training data, the Bayesian DNN is also more robust than a DNN. We further implement the system on a UR10 robot, and test the robustness of the Bayesian DNN and the IBVS convergence speed. Results show that the Bayesian DNN can avoid the poses out of the reach range of the robot and it lets the IBVS task converge faster than the opposite repulsive pose.1

show abstract

Section: Related Workmentioning

confidence: 99%

A Bayesian Deep Neural Network for Safe Visual Servoing in Human–Robot Interaction

2021

View full text Add to dashboard Cite

show abstract

“…A. Johnston et al [47] presented self-attention and discrete disparity prediction to improve the self-supervised monocular trained depth estimation. M. Poggi et al [48] explored for the first time how to estimate the uncertainty for the self-supervised paradigms for monocular depth estimation task and how this effect depth accuracy.…”

Section: Unsupervised Depth Estimationmentioning

confidence: 99%

Encoder-Decoder Structure With the Feature Pyramid for Depth Estimation From a Single Image

et al. 2021

View full text Add to dashboard Cite

We address the problem of depth estimation from a single monocular image in the paper. Depth estimation from a single image is an ill-posed and inherently ambiguous problem. In the paper, we propose an encoder-decoder structure with the feature pyramid to predict the depth map from a single RGB image. More specifically, the feature pyramid is used to detect objects of different scales in the image. The encoder structure aims to extract the most representative information from the original image through a series of convolution operations and to reduce the resolution of the input image. We adopt Res2-50 as the encoder to extract important features. The decoder section uses a novel upsampling structure to improve the output resolution. Then, we also propose a novel loss function that adds gradient loss and surface normal loss to the depth loss, which can predict not only the global depth but also the depth of fuzzy edges and small objects. Additionally, we use Adam as our optimization function to optimize our network and speed up convergence. Our extensive experimental evaluation proves the efficiency and effectiveness of the method, which is competitive with previous methods on the Make3D dataset and outperforms state-of-theart methods on the NYU Depth v2 dataset.

show abstract

“…In all of these probabilistic methods, only a unimodal distribution is used to model the aleatoric uncertainty, considering the error distribution as a Gaussian (Kendall and Gal, 2017) or a Laplacian distribution (Poggi et al, 2020;Mehltretter and Heipke, 2021). However, this is not always the case in the context of dense stereo matching, especially in real-world scenarios this assumption is violated by outlier measurements or by commonly challenging regions, such as occlusion, texture-less regions and depth discontinuities.…”

Section: Related Workmentioning

confidence: 99%

Mixed Probability Models for Aleatoric Uncertainty Estimation in the Context of Dense Stereo Matching

Zhang

Mehltretter

2021

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. The ability to identify erroneous depth estimates is of fundamental interest. Information regarding the aleatoric uncertainty of depth estimates can be, for example, used to support the process of depth reconstruction itself. Consequently, various methods for the estimation of aleatoric uncertainty in the context of dense stereo matching have been presented in recent years, with deep learning-based approaches being particularly popular. Among these deep learning-based methods, probabilistic strategies are increasingly attracting interest, because the estimated uncertainty can be quantified in pixels or in metric units due to the consideration of real error distributions. However, existing probabilistic methods usually assume a unimodal distribution to describe the error distribution while simply neglecting cases in real-world scenarios that could violate this assumption. To overcome this limitation, we propose two novel mixed probability models consisting of Laplacian and Uniform distributions for the task of aleatoric uncertainty estimation. In this way, we explicitly address commonly challenging regions in the context of dense stereo matching and outlier measurements, respectively. To allow a fair comparison, we adapt a common neural network architecture to investigate the effects of the different uncertainty models. In an extensive evaluation using two datasets and two common dense stereo matching methods, the proposed methods demonstrate state-of-the-art accuracy.

show abstract

On the Uncertainty of Self-Supervised Monocular Depth Estimation

Cited by 210 publications

References 54 publications

A Bayesian Deep Neural Network for Safe Visual Servoing in Human–Robot Interaction

A Bayesian Deep Neural Network for Safe Visual Servoing in Human–Robot Interaction

Encoder-Decoder Structure With the Feature Pyramid for Depth Estimation From a Single Image

Mixed Probability Models for Aleatoric Uncertainty Estimation in the Context of Dense Stereo Matching

Contact Info

Product

Resources

About