LongSplat: Robust Unposed 3D Gaussian Splatting for Casual Long Videos

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Chin-Yang, Sun, Cheng, Yang, Fu-En, Chen, Min-Hung, Lin, Yen-Yu, Liu, Yu-Lun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909743735373824
author Lin, Chin-Yang
Sun, Cheng
Yang, Fu-En
Chen, Min-Hung
Lin, Yen-Yu
Liu, Yu-Lun
author_facet Lin, Chin-Yang
Sun, Cheng
Yang, Fu-En
Chen, Min-Hung
Lin, Yen-Yu
Liu, Yu-Lun
contents 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/
format Preprint
id arxiv_https___arxiv_org_abs_2508_14041
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LongSplat: Robust Unposed 3D Gaussian Splatting for Casual Long Videos
Lin, Chin-Yang
Sun, Cheng
Yang, Fu-En
Chen, Min-Hung
Lin, Yen-Yu
Liu, Yu-Lun
Computer Vision and Pattern Recognition
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/
title LongSplat: Robust Unposed 3D Gaussian Splatting for Casual Long Videos
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.14041