Journal
COMPUTER VISION - ECCV 2022, PT II
Volume 13662, Issue -, Pages 237-256Publisher
SPRINGER INTERNATIONAL PUBLISHING AG
DOI: 10.1007/978-3-031-20086-1_14
Keywords
Monocular depth estimation; Frequentist uncertainty quantification; Constrained ordinal regression; Bootstrapping
Funding
- LIEF Grant [LE170100200]
- NCI Adapter Scheme
- Australian Government
- ARC [DE210101624]
- Research Computing Services NCI Access scheme at The University of Melbourne
- Australian Research Council [DE210101624] Funding Source: Australian Research Council
Ask authors/readers for more resources
This paper presents an uncertainty quantification method for supervised MDE models, capturing uncertainty through predictive variance and estimating error variance and estimation variance using constrained ordinal regression and bootstrapping methods. Experimental results demonstrate the accuracy and effectiveness of the proposed method.
Monocular Depth Estimation (MDE) is a task to predict a dense depth map from a single image. Despite the recent progress brought by deep learning, existing methods are still prone to errors due to the ill-posed nature of MDE. Hence depth estimation systems must be self-aware of possible mistakes to avoid disastrous consequences. This paper provides an uncertainty quantification method for supervised MDE models. From a frequentist view, we capture the uncertainty by predictive variance that consists of two terms: error variance and estimation variance. The former represents the noise of a depth value, and the latter measures the randomness in the depth regression model due to training on finite data. To estimate error variance, we perform constrained ordinal regression (ConOR) on discretized depth to estimate the conditional distribution of depth given image, and then compute the corresponding conditional mean and variance as the predicted depth and error variance estimator, respectively. Our work also leverages bootstrapping methods to infer estimation variance from re-sampled data. We perform experiments on both simulated and real data to validate the effectiveness of the proposed method. The results show that our approach produces accurate uncertainty estimates while maintaining high depth prediction accuracy. The code is available at https://github.com/timmy11hu/ConOR
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available