Revisiting Gradient-based Uncertainty for Monocular Depth Estimation

TL;DR

This paper introduces a simple, post hoc gradient-based method to assess uncertainty in monocular depth estimation models without retraining, using an auxiliary loss and back-propagation on feature maps, validated on KITTI and NYU datasets.

Contribution

It proposes a novel, effective gradient-based uncertainty estimation technique that works on pre-trained models using an auxiliary loss and back-propagation, without needing ground truth during inference.

Findings

Outperforms related uncertainty methods on KITTI and NYU benchmarks.

Effective for models trained on monocular sequences prone to uncertainty.

Provides publicly available code and models for reproducibility.

Abstract

Monocular depth estimation, similar to other image-based tasks, is prone to erroneous predictions due to ambiguities in the image, for example, caused by dynamic objects or shadows. For this reason, pixel-wise uncertainty assessment is required for safety-critical applications to highlight the areas where the prediction is unreliable. We address this in a post hoc manner and introduce gradient-based uncertainty estimation for already trained depth estimation models. To extract gradients without depending on the ground truth depth, we introduce an auxiliary loss function based on the consistency of the predicted depth and a reference depth. The reference depth, which acts as pseudo ground truth, is in fact generated using a simple image or feature augmentation, making our approach simple and effective. To obtain the final uncertainty score, the derivatives w.r.t. the feature maps from single or multiple layers are calculated using back-propagation. We demonstrate that our gradient-based approach is effective in determining the uncertainty without re-training using the two standard depth estimation benchmarks KITTI and NYU. In particular, for models trained with monocular sequences and therefore most prone to uncertainty, our method outperforms related approaches. In addition, we publicly provide our code and models: https://github.com/jhornauer/GrUMoDepth