We propose Long-LRM, a feed-forward 3D Gaussian reconstruction model for instant, high-resolution, 360° wide-coverage, scene-level reconstruction. Specifically, it takes in 32 input images at a resolution of 960×540 and produces the Gaussian reconstruction in just 1 second on a single A100 GPU. To handle the long sequence of 250K tokens brought by the large input size, Long-LRM features a mixture of the recent Mamba2 blocks and the classical transformer blocks, enhanced by a light-weight token merging module and Gaussian pruning steps that balance between quality and efficiency. We evaluate Long-LRM on the large-scale DL3DV benchmark and Tanks&Temples, demonstrating reconstruction quality comparable to the optimization-based methods while achieving an 800× speedup w.r.t. the optimization-based approaches and an input size at least 60× larger than the previous feed-forward approaches. We conduct extensive ablation studies on our model design choices for both rendering quality and computation efficiency. We also explore Long-LRM's compatibility with other Gaussian variants such as 2D GS, which enhances Long-LRM's ability in geometry reconstruction.
@inproceedings{ziwen2025llrm,
title={Long-LRM: Long-sequence Large Reconstruction Model for Wide-coverage Gaussian Splats},
author={Ziwen, Chen and Tan, Hao and Zhang, Kai and Bi, Sai and Luan, Fujun and Hong, Yicong and Fuxin, Li and Xu, Zexiang},
booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
year={2025}
}