Depth Any Panoramas: A Foundation Model for Panoramic Depth Estimation

TL;DR

This paper introduces a panoramic depth estimation foundation model that generalizes well across various scenes by leveraging a large, diverse dataset and novel training strategies, achieving strong zero-shot performance.

Contribution

The work presents a new panoramic depth model with a comprehensive data collection pipeline and innovative training techniques for improved robustness and generalization.

Findings

Strong performance on multiple benchmarks

Effective zero-shot generalization to real-world scenes

Robust and stable depth predictions across diverse environments

Abstract

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/}