构建全景深度基础模型，跨场景距离实现高精度零样本预测。
Depth Any Panoramas: A Foundation Model for Panoramic Depth Estimation
- 融合真实、合成与文本生成数据，构建大规模全景深度训练集。
- 在多个基准上实现零样本泛化，真实场景下深度预测稳定可靠。
- 提出可插拔距离掩码头与几何优化机制，增强远近场景适应性。
本文提出一个全景度量深度基础模型，可在不同场景距离间实现良好泛化。我们采用数据驱动的闭环范式，在数据构建与框架设计两方面进行探索。通过整合公开数据集、自建UE5模拟器生成的高质量合成数据、文本生成图像模型产出的数据，以及网络获取的真实全景图像，构建大规模数据集。为减少室内/室外、合成/真实数据间的领域差异，引入三阶段伪标签清洗流程，为未标注图像生成可靠真值。模型采用DINOv3-Large作为主干网络以利用其强泛化能力，并设计可插拔的距离掩码头、基于清晰度的优化策略及基于几何一致性的优化方法，提升对不同距离变化的鲁棒性。在Stanford2D3D、Matterport3D和Deep360等多个基准上的实验表明，该模型具备优异性能与零样本泛化能力，尤其在多样化真实场景中表现出稳定且精确的度量深度预测。
原文摘要 · Abstract (English)
In this work, we present a panoramic metric depth foundation model that generalizes across diverse scene distances. We explore a data-in-the-loop paradigm from the view of both data construction and framework design. We collect a large-scale dataset by combining public datasets, high-quality synthetic data from our UE5 simulator and text-to-image models, and real panoramic images from the web. To reduce domain gaps between indoor/outdoor and synthetic/real data, we introduce a three-stage pseudo-label curation pipeline to generate reliable ground truth for unlabeled images. For the model, we adopt DINOv3-Large as the backbone for its strong pre-trained generalization, and introduce a plug-and-play range mask head, sharpness-centric optimization, and geometry-centric optimization to improve robustness to varying distances and enforce geometric consistency across views. Experiments on multiple benchmarks (e.g., Stanford2D3D, Matterport3D, and Deep360) demonstrate strong performance and zero-shot generalization, with particularly robust and stable metric predictions in diverse real-world scenes. The project page can be found at: \href{https://insta360-research-team.github.io/DAP_website/} {https://insta360-research-team.github.io/DAP\_website/}
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。