Splatonic: Architecture Support for 3D Gaussian Splatting SLAM via Sparse Processing
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_ | 1866917107347750912 |
|---|---|
| author | Huang, Xiaotong Zhu, He Ma, Tianrui Xiong, Yuxiang Liu, Fangxin He, Zhezhi Gan, Yiming Liu, Zihan Leng, Jingwen Feng, Yu Guo, Minyi |
| author_facet | Huang, Xiaotong Zhu, He Ma, Tianrui Xiong, Yuxiang Liu, Fangxin He, Zhezhi Gan, Yiming Liu, Zihan Leng, Jingwen Feng, Yu Guo, Minyi |
| contents | 3D Gaussian splatting (3DGS) has emerged as a promising direction for SLAM due to its high-fidelity reconstruction and rapid convergence. However, 3DGS-SLAM algorithms remain impractical for mobile platforms due to their high computational cost, especially for their tracking process.
This work introduces Splatonic, a sparse and efficient real-time 3DGS-SLAM algorithm-hardware co-design for resource-constrained devices. Inspired by classical SLAMs, we propose an adaptive sparse pixel sampling algorithm that reduces the number of rendered pixels by up to 256$\times$ while retaining accuracy. To unlock this performance potential on mobile GPUs, we design a novel pixel-based rendering pipeline that improves hardware utilization via Gaussian-parallel rendering and preemptive $α$-checking. Together, these optimizations yield up to 121.7$\times$ speedup on the bottleneck stages and 14.6$\times$ end-to-end speedup on off-the-shelf GPUs. To further address new bottlenecks introduced by our rendering pipeline, we propose a pipelined architecture that simplifies the overall design while addressing newly emerged bottlenecks in projection and aggregation. Evaluated across four 3DGS-SLAM algorithms, Splatonic achieves up to 274.9$\times$ speedup and 4738.5$\times$ energy savings over mobile GPUs and up to 25.2$\times$ speedup and 241.1$\times$ energy savings over state-of-the-art accelerators, all with comparable accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_18755 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Splatonic: Architecture Support for 3D Gaussian Splatting SLAM via Sparse Processing Huang, Xiaotong Zhu, He Ma, Tianrui Xiong, Yuxiang Liu, Fangxin He, Zhezhi Gan, Yiming Liu, Zihan Leng, Jingwen Feng, Yu Guo, Minyi Hardware Architecture 3D Gaussian splatting (3DGS) has emerged as a promising direction for SLAM due to its high-fidelity reconstruction and rapid convergence. However, 3DGS-SLAM algorithms remain impractical for mobile platforms due to their high computational cost, especially for their tracking process. This work introduces Splatonic, a sparse and efficient real-time 3DGS-SLAM algorithm-hardware co-design for resource-constrained devices. Inspired by classical SLAMs, we propose an adaptive sparse pixel sampling algorithm that reduces the number of rendered pixels by up to 256$\times$ while retaining accuracy. To unlock this performance potential on mobile GPUs, we design a novel pixel-based rendering pipeline that improves hardware utilization via Gaussian-parallel rendering and preemptive $α$-checking. Together, these optimizations yield up to 121.7$\times$ speedup on the bottleneck stages and 14.6$\times$ end-to-end speedup on off-the-shelf GPUs. To further address new bottlenecks introduced by our rendering pipeline, we propose a pipelined architecture that simplifies the overall design while addressing newly emerged bottlenecks in projection and aggregation. Evaluated across four 3DGS-SLAM algorithms, Splatonic achieves up to 274.9$\times$ speedup and 4738.5$\times$ energy savings over mobile GPUs and up to 25.2$\times$ speedup and 241.1$\times$ energy savings over state-of-the-art accelerators, all with comparable accuracy. |
| title | Splatonic: Architecture Support for 3D Gaussian Splatting SLAM via Sparse Processing |
| topic | Hardware Architecture |
| url | https://arxiv.org/abs/2511.18755 |