Boosting Omnidirectional Stereo Matching with a Pre-trained Depth Foundation Model

-- Omnidirectional depth perception is essential for mobile robotics applications that require scene understanding across a full 360 field of view. Camera-based setups offer a cost-effective option by using stereo depth estimation to generate dense, high-resolution depth maps without relying on expensive active sensing. However, existing omnidirectional stereo matching approaches achieve only limited depth accuracy across diverse environments, depth ranges, and lighting conditions, due to the scarcity of real-world data. We introduce a dedicated two-stage training strategy to utilize the relative monocular depth features for our omnidirectional stereo matching before scale-invariant fine-tuning. DFI-OmniStereo achieves state-of-the-art results on the real-world Helvipad dataset, reducing disparity MAE by approximately 16% compared to the previous best omnidirectional stereo method. I. INTRODUCTION Mobile robots are increasingly being deployed across various domains, including agriculture [1], autonomous driving [2], healthcare [3], search and rescue missions [4], and warehouse automation [5]. In these applications, accurate depth perception is crucial to construct reliable 3D representations of a robot's environment to achieve essential tasks such as path planning, mapping, and manipulation. Traditionally, LiDAR sensors have been the preferred choice for acquiring depth information due to their high precision and 360 field of view.