PanSplat: 4K Panorama Synthesis with Feed-Forward Gaussian Splatting
This addresses the need for high-resolution, fast, and memory-efficient panorama synthesis in applications like VR and robotics, representing a strong specific gain rather than a foundational advancement.
The paper tackles the problem of high-resolution panorama view synthesis, which is constrained to lower resolutions by existing methods, and presents PanSplat, achieving state-of-the-art results with support for 4K resolution and efficient training on a single GPU.
With the advent of portable 360° cameras, panorama has gained significant attention in applications like virtual reality (VR), virtual tours, robotics, and autonomous driving. As a result, wide-baseline panorama view synthesis has emerged as a vital task, where high resolution, fast inference, and memory efficiency are essential. Nevertheless, existing methods are typically constrained to lower resolutions (512 $\times$ 1024) due to demanding memory and computational requirements. In this paper, we present PanSplat, a generalizable, feed-forward approach that efficiently supports resolution up to 4K (2048 $\times$ 4096). Our approach features a tailored spherical 3D Gaussian pyramid with a Fibonacci lattice arrangement, enhancing image quality while reducing information redundancy. To accommodate the demands of high resolution, we propose a pipeline that integrates a hierarchical spherical cost volume and Gaussian heads with local operations, enabling two-step deferred backpropagation for memory-efficient training on a single A100 GPU. Experiments demonstrate that PanSplat achieves state-of-the-art results with superior efficiency and image quality across both synthetic and real-world datasets. Code is available at https://github.com/chengzhag/PanSplat.