RoSAMDepth：利用分割一切模型的鲁棒自监督深度估计
RoSAMDepth: Robust Self-supervised Depth Estimation Leveraging Segment Anything Model
把 SAM 的物体级分割先验注入自监督深度估计：用分割引导的特征对比、区域高斯似然损失和物体级可靠性估计来抑制物体内部深度漂移与恶劣天气下的模糊，可直接复用到单目深度/自监督训练框架中。
原文依据 a segment-guided representation contrasting method that injects object-level awareness into the feature representation space
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论鲁棒深度估计旨在在多样条件下保持高质量深度。然而，现有大多数方法在估计深度时未考虑物体级信息。因此，预测深度容易在物体内部发生偏离，并在恶劣条件下变得模糊。为克服这一缺陷，我们提出 RoSAMDepth，一种新颖框架，可辅助鲁棒自监督深度估计利用来自分割一切模型（SAM）的丰富多样的物体级先验。我们聚焦于在三个关键方面融入物体级信息：一种分割引导的表征对比方法，将物体级感知注入特征表示空间；一种自适应区域异常值掩码策略，结合区域高斯似然损失以强制区域深度平滑；以及一种物体级可靠性估计策略，以减轻不可靠监督的影响。在多个数据集和多样天气条件下的广泛实验表明，我们的方法产生更清晰、更准确的深度预测，持续优于最优方法。
英文原文摘要 1175 字符
Robust depth estimation aims to maintain high-quality depths across diverse conditions. However, most existing methods estimate depth without taking into account the object-level information. As a result, the predicted depth may easily deviate within objects and become blurred under adverse conditions. To overcome this weakness, we propose RoSAMDepth, a novel framework that can assist robust self-supervised depth estimation in leveraging rich and diverse object-level priors from the Segment Anything Model (SAM). We focus on incorporating object-level information across three key aspects: a segment-guided representation contrasting method that injects object-level awareness into the feature representation space; an adaptive regional outlier masking strategy combined with a regional Gaussian likelihood loss that enforces regional depth smoothness; and an object-level reliability estimation strategy that mitigates the influence of unreliable supervision. Extensive experiments across multiple datasets and diverse weather conditions demonstrate that our method produces sharper, more accurate depth predictions, consistently outperforming state-of-the-art methods.