Shedding Light on Depth: Explainability Assessment in Monocular Depth Estimation

2025-09-19

AI Summary

Monocular depth estimation (MDE) models suffer from a lack of interpretable evaluation, hindering their trustworthy deployment in safety-critical applications. This work systematically investigates the decision attribution mechanisms underlying MDE networks—from input images to predicted depth maps—and introduces Attribution Fidelity, the first metric to quantitatively measure consistency between attribution maps and depth predictions. It effectively identifies unreliable explanations overlooked by conventional metrics such as Infidelity. We empirically evaluate Saliency Maps, Integrated Gradients, and Attention Rollout across both lightweight and deep MDE architectures: Saliency Maps achieve superior performance on lightweight models, while Attention Rollout demonstrates greater robustness on deeper ones. Results show that Attribution Fidelity significantly improves the accuracy of reliability assessment for attribution methods, establishing a reproducible, comparable benchmark framework for explainability analysis of MDE models.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Explainable artificial intelligence is increasingly employed to understand the decision-making process of deep learning models and create trustworthiness in their adoption. However, the explainability of Monocular Depth Estimation (MDE) remains largely unexplored despite its wide deployment in real-world applications. In this work, we study how to analyze MDE networks to map the input image to the predicted depth map. More in detail, we investigate well-established feature attribution methods, Saliency Maps, Integrated Gradients, and Attention Rollout on different computationally complex models for MDE: METER, a lightweight network, and PixelFormer, a deep network. We assess the quality of the generated visual explanations by selectively perturbing the most relevant and irrelevant pixels, as identified by the explainability methods, and analyzing the impact of these perturbations on the model's output. Moreover, since existing evaluation metrics can have some limitations in measuring the validity of visual explanations for MDE, we additionally introduce the Attribution Fidelity. This metric evaluates the reliability of the feature attribution by assessing their consistency with the predicted depth map. Experimental results demonstrate that Saliency Maps and Integrated Gradients have good performance in highlighting the most important input features for MDE lightweight and deep models, respectively. Furthermore, we show that Attribution Fidelity effectively identifies whether an explainability method fails to produce reliable visual maps, even in scenarios where conventional metrics might suggest satisfactory results.

Problem

• Assessing explainability methods for monocular depth estimation models
• Evaluating visual explanation quality through selective pixel perturbation
• Introducing Attribution Fidelity metric to validate feature attribution reliability

Innovation

• Feature attribution methods for depth estimation
• Attribution Fidelity metric for explanation evaluation
• Selective pixel perturbation analysis technique