LongSplat: Robust Unposed 3D Gaussian Splatting for Casual Long Videos
It addresses a domain-specific problem for 3D reconstruction and rendering from casual videos, with incremental improvements over existing methods.
LongSplat tackles novel view synthesis from casually captured long videos with irregular camera motion and unknown poses, achieving state-of-the-art results by improving rendering quality, pose accuracy, and computational efficiency.
LongSplat addresses critical challenges in novel view synthesis (NVS) from casually captured long videos characterized by irregular camera motion, unknown camera poses, and expansive scenes. Current methods often suffer from pose drift, inaccurate geometry initialization, and severe memory limitations. To address these issues, we introduce LongSplat, a robust unposed 3D Gaussian Splatting framework featuring: (1) Incremental Joint Optimization that concurrently optimizes camera poses and 3D Gaussians to avoid local minima and ensure global consistency; (2) a robust Pose Estimation Module leveraging learned 3D priors; and (3) an efficient Octree Anchor Formation mechanism that converts dense point clouds into anchors based on spatial density. Extensive experiments on challenging benchmarks demonstrate that LongSplat achieves state-of-the-art results, substantially improving rendering quality, pose accuracy, and computational efficiency compared to prior approaches. Project page: https://linjohnss.github.io/longsplat/