Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.08334 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917360485531648 |
|---|---|
| author | Liu, Chang Yuan, Hongliang Zhang, Lianghao Wang, Sichao Guo, Jianwei Huang, Shi-Sheng |
| author_facet | Liu, Chang Yuan, Hongliang Zhang, Lianghao Wang, Sichao Guo, Jianwei Huang, Shi-Sheng |
| contents | Rendering complex reflection of real-world scenes using 3D Gaussian splatting has been a quite promising solution for photorealistic novel view synthesis, but still faces bottlenecks especially in rendering speed and memory storage. This paper proposes a new Hybrid Splatting(HybridSplat) mechanism for Gaussian primitives. Our key idea is a new reflection-baked Gaussian tracing, which bakes the view-dependent reflection within each Gaussian primitive while rendering the reflection using tile-based Gaussian splatting. Then we integrate the reflective Gaussian primitives with base Gaussian primitives using a unified hybrid splatting framework for high-fidelity scene reconstruction. Moreover, we further introduce a pipeline-level acceleration for the hybrid splatting, and reflection-sensitive Gaussian pruning to reduce the model size, thus achieving much faster rendering speed and lower memory storage while preserving the reflection rendering quality. By extensive evaluation, our HybridSplat accelerates about 7x rendering speed across complex reflective scenes from Ref-NeRF, NeRF-Casting with 4x fewer Gaussian primitives than similar ray-tracing based Gaussian splatting baselines, serving as a new state-of-the-art method especially for complex reflective scenes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_08334 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | HybridSplat: Fast Reflection-baked Gaussian Tracing using Hybrid Splatting Liu, Chang Yuan, Hongliang Zhang, Lianghao Wang, Sichao Guo, Jianwei Huang, Shi-Sheng Computer Vision and Pattern Recognition Rendering complex reflection of real-world scenes using 3D Gaussian splatting has been a quite promising solution for photorealistic novel view synthesis, but still faces bottlenecks especially in rendering speed and memory storage. This paper proposes a new Hybrid Splatting(HybridSplat) mechanism for Gaussian primitives. Our key idea is a new reflection-baked Gaussian tracing, which bakes the view-dependent reflection within each Gaussian primitive while rendering the reflection using tile-based Gaussian splatting. Then we integrate the reflective Gaussian primitives with base Gaussian primitives using a unified hybrid splatting framework for high-fidelity scene reconstruction. Moreover, we further introduce a pipeline-level acceleration for the hybrid splatting, and reflection-sensitive Gaussian pruning to reduce the model size, thus achieving much faster rendering speed and lower memory storage while preserving the reflection rendering quality. By extensive evaluation, our HybridSplat accelerates about 7x rendering speed across complex reflective scenes from Ref-NeRF, NeRF-Casting with 4x fewer Gaussian primitives than similar ray-tracing based Gaussian splatting baselines, serving as a new state-of-the-art method especially for complex reflective scenes. |
| title | HybridSplat: Fast Reflection-baked Gaussian Tracing using Hybrid Splatting |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.08334 |