Region-aware Depth Scale Adaptation with Sparse Measurements

TL;DR

This paper presents a non-learning-based method that uses sparse measurements to convert relative depth predictions from foundation models into metric scale, maintaining generalization without retraining.

Contribution

It introduces a novel approach that adapts depth predictions to metric scale using sparse data without retraining or fine-tuning, preserving model generalization.

Findings

Effective in converting relative to metric depth

No retraining or fine-tuning required

Maintains original model generalization

Abstract

In recent years, the emergence of foundation models for depth prediction has led to remarkable progress, particularly in zero-shot monocular depth estimation. These models generate impressive depth predictions; however, their outputs are often in relative scale rather than metric scale. This limitation poses challenges for direct deployment in real-world applications. To address this, several scale adaptation methods have been proposed to enable foundation models to produce metric depth. However, these methods are typically costly, as they require additional training on new domains and datasets. Moreover, fine-tuning these models often compromises their original generalization capabilities, limiting their adaptability across diverse scenes. In this paper, we introduce a non-learning-based approach that leverages sparse depth measurements to adapt the relative-scale predictions of foundation models into metric-scale depth. Our method requires neither retraining nor fine-tuning, thereby preserving the strong generalization ability of the original foundation models while enabling them to produce metric depth. Experimental results demonstrate the effectiveness of our approach, high-lighting its potential to bridge the gap between relative and metric depth without incurring additional computational costs or sacrificing generalization ability.