FUSER: Feed-Forward MUltiview 3D Registration Transformer and SE(3)$^N$ Diffusion Refinement
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910168144412672 |
|---|---|
| author | Jiang, Haobo Xie, Jin Yang, Jian Yu, Liang Zheng, Jianmin |
| author_facet | Jiang, Haobo Xie, Jin Yang, Jian Yu, Liang Zheng, Jianmin |
| contents | Registration of multiview point clouds conventionally relies on extensive pairwise matching to build a pose graph for global synchronization, which is computationally expensive and inherently ill-posed without holistic geometric constraints. This paper proposes FUSER, the first feed-forward multiview registration transformer that jointly processes all scans in a unified, compact latent space to directly predict global poses without any pairwise estimation. To maintain tractability, FUSER encodes each scan into low-resolution superpoint features via a sparse 3D CNN that preserves absolute translation cues, and performs efficient intra- and inter-scan reasoning through a Geometric Alternating Attention module. Particularly, we transfer 2D attention priors from off-the-shelf foundation models to enhance 3D feature interaction and geometric consistency. Building upon FUSER, we further introduce FUSER-DF, an SE(3)$^N$ diffusion refinement framework to correct FUSER's estimates via denoising in the joint SE(3)$^N$ space. FUSER acts as a surrogate multiview registration model to construct the denoiser, and a prior-conditioned SE(3)$^N$ variational lower bound is derived for denoising supervision. Extensive experiments on 3DMatch, ScanNet and ArkitScenes demonstrate that our approach achieves the superior registration accuracy and outstanding computational efficiency. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_09373 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FUSER: Feed-Forward MUltiview 3D Registration Transformer and SE(3)$^N$ Diffusion Refinement Jiang, Haobo Xie, Jin Yang, Jian Yu, Liang Zheng, Jianmin Computer Vision and Pattern Recognition Registration of multiview point clouds conventionally relies on extensive pairwise matching to build a pose graph for global synchronization, which is computationally expensive and inherently ill-posed without holistic geometric constraints. This paper proposes FUSER, the first feed-forward multiview registration transformer that jointly processes all scans in a unified, compact latent space to directly predict global poses without any pairwise estimation. To maintain tractability, FUSER encodes each scan into low-resolution superpoint features via a sparse 3D CNN that preserves absolute translation cues, and performs efficient intra- and inter-scan reasoning through a Geometric Alternating Attention module. Particularly, we transfer 2D attention priors from off-the-shelf foundation models to enhance 3D feature interaction and geometric consistency. Building upon FUSER, we further introduce FUSER-DF, an SE(3)$^N$ diffusion refinement framework to correct FUSER's estimates via denoising in the joint SE(3)$^N$ space. FUSER acts as a surrogate multiview registration model to construct the denoiser, and a prior-conditioned SE(3)$^N$ variational lower bound is derived for denoising supervision. Extensive experiments on 3DMatch, ScanNet and ArkitScenes demonstrate that our approach achieves the superior registration accuracy and outstanding computational efficiency. |
| title | FUSER: Feed-Forward MUltiview 3D Registration Transformer and SE(3)$^N$ Diffusion Refinement |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.09373 |