High-Resolution Depth Estimation for 360-degree Panoramas through Perspective and Panoramic Depth Images Registration
This work addresses the need for efficient and accurate depth estimation in panoramic imaging, which is incremental as it builds on existing stitching approaches but simplifies the process.
The paper tackles the problem of generating high-resolution depth maps for 360-degree panoramas by proposing a method that registers perspective depth images to an existing panoramic depth map, resulting in faster computation and improved accuracy compared to the state-of-the-art 360MonoDepth, with outputs up to 2048x1024 resolution.
We propose a novel approach to compute high-resolution (2048x1024 and higher) depths for panoramas that is significantly faster and qualitatively and qualitatively more accurate than the current state-of-the-art method (360MonoDepth). As traditional neural network-based methods have limitations in the output image sizes (up to 1024x512) due to GPU memory constraints, both 360MonoDepth and our method rely on stitching multiple perspective disparity or depth images to come out a unified panoramic depth map. However, to achieve globally consistent stitching, 360MonoDepth relied on solving extensive disparity map alignment and Poisson-based blending problems, leading to high computation time. Instead, we propose to use an existing panoramic depth map (computed in real-time by any panorama-based method) as the common target for the individual perspective depth maps to register to. This key idea made producing globally consistent stitching results from a straightforward task. Our experiments show that our method generates qualitatively better results than existing panorama-based methods, and further outperforms them quantitatively on datasets unseen by these methods.