Distill Any Depth: Distillation Creates a Stronger Monocular Depth Estimator
This addresses a specific bottleneck in zero-shot monocular depth estimation for computer vision applications, representing an incremental improvement over existing distillation methods.
The paper tackles the problem of noise amplification in pseudo-label distillation for monocular depth estimation by showing that depth normalization isn't always necessary and proposing Cross-Context Distillation with assistant-guided strategies. The result is a method that significantly outperforms state-of-the-art approaches on benchmark datasets.
Recent advances in zero-shot monocular depth estimation(MDE) have significantly improved generalization by unifying depth distributions through normalized depth representations and by leveraging large-scale unlabeled data via pseudo-label distillation. However, existing methods that rely on global depth normalization treat all depth values equally, which can amplify noise in pseudo-labels and reduce distillation effectiveness. In this paper, we present a systematic analysis of depth normalization strategies in the context of pseudo-label distillation. Our study shows that, under recent distillation paradigms (e.g., shared-context distillation), normalization is not always necessary, as omitting it can help mitigate the impact of noisy supervision. Furthermore, rather than focusing solely on how depth information is represented, we propose Cross-Context Distillation, which integrates both global and local depth cues to enhance pseudo-label quality. We also introduce an assistant-guided distillation strategy that incorporates complementary depth priors from a diffusion-based teacher model, enhancing supervision diversity and robustness. Extensive experiments on benchmark datasets demonstrate that our approach significantly outperforms state-of-the-art methods, both quantitatively and qualitatively.