Accuracy Does Not Guarantee Human-Likeness in Monocular Depth Estimators
This work highlights a key challenge for autonomous driving and robotics by showing that current benchmarks may not align with human perception, suggesting the need for more human-centric evaluations.
The study investigated the relationship between model accuracy and human similarity in monocular depth estimation, analyzing 69 estimators on the KITTI dataset, and found that improving accuracy does not necessarily lead to more human-like behavior, with distinct trade-offs observed.
Monocular depth estimation is a fundamental capability for real-world applications such as autonomous driving and robotics. Although deep neural networks (DNNs) have achieved superhuman accuracy on physical-based benchmarks, a key challenge remains: aligning model representations with human perception, a promising strategy for enhancing model robustness and interpretability. Research in object recognition has revealed a complex trade-off between model accuracy and human-like behavior, raising a question whether a similar divergence exist in depth estimation, particularly for natural outdoor scenes where benchmarks rely on sensor-based ground truth rather than human perceptual estimates. In this study, we systematically investigated the relationship between model accuracy and human similarity across 69 monocular depth estimators using the KITTI dataset. To dissect the structure of error patterns on a factor-by-factor basis, we applied affine fitting to decompose prediction errors into interpretable components. Intriguingly, our results reveal while humans and DNNs share certain estimation biases (positive error correlations), we observed distinct trade-off relationships between model accuracy and human similarity. This finding indicates that improving accuracy does not necessarily lead to more human-like behavior, underscoring the necessity of developing multifaceted, human-centric evaluations beyond traditional accuracy.