StereoSplat+: Feed-Forward Stereo Gaussian Splatting with Diffusion-Assisted Progressive Inference
Quick Answer
This paper shows that StereoSplat+ introduces a diffusion-enhanced feed-forward framework for causal 3D Gaussian Splatting from a single stereo pair, improving rendering quality and geometry accuracy in occluded regions.
Quick Take
StereoSplat+ introduces a diffusion-enhanced feed-forward framework for causal 3D Gaussian Splatting from a single stereo pair, improving rendering quality and geometry accuracy in occluded regions. It outperforms existing methods on the KITTI-360 dataset, addressing challenges in robotics and AR where multi-view data is unavailable.
Key Points
- StereoSplat+ enables high-quality 3D scene reconstruction from a single stereo observation.
- It incorporates a cost-volume branch and a triplane-based 3D volume branch for geometry fusion.
- The method enhances rendering quality, particularly in occluded areas and under strong view extrapolation.
- Experiments on the KITTI-360 dataset demonstrate significant performance improvements over existing baselines.
- Accepted for presentation at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2026).
Paper Resources
📖 Reader Mode
~2 min readAbstract:Recent advances in 3D Gaussian Splatting (3DGS) have enabled high-quality, render-ready scene representations for novel-view synthesis. However, most existing 3DGS pipelines rely on multi-view observations (or non-causal access to future frames) to achieve sufficient coverage, which is often unavailable in on-device robotics and AR settings where sensing is restricted to a single stereo rig. Recovering a high-quality 3DGS scene from one stereo observation, therefore, remains challenging due to occlusions, limited field of view, and missing geometry. We present StereoSplat+, a diffusion-enhanced feed-forward framework that enables causal reconstruction from a single stereo pair. Our method builds on two key components. First, we propose StereoSplat, an input-invariant feed-forward 3D Gaussian estimator that takes a variable number of posed stereo pairs as input and predicts high-quality 3D Gaussians. StereoSplat fuses complementary geometry cues via a cost-volume branch and a triplane-based 3D volume branch and leverages continuous pose encoding to generalize across view counts and camera configurations. Second, since multiple posed stereo pairs are typically unavailable at inference time, we introduce a diffusion-enhanced one-shot progressive inference scheme called StereoSplat+: starting from one stereo pair, we render novel stereo views from the predicted 3DGS, refine them with a one-step diffusion enhancer, and feed them back as additional inputs to update the 3DGS. Experiments on the KITTI-360 dataset show that StereoSplat+ improves novel-view rendering quality and geometry accuracy, especially in occluded regions and under strong view extrapolation, outperforming recent feed-forward 3DGS baselines.
| Comments: | 8 pages, accepted as a conference paper for IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS2026) |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2607.08808 [cs.CV] |
| (or arXiv:2607.08808v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.08808 arXiv-issued DOI via DataCite |
Submission history
From: Zihua Liu [view email]
[v1]
Thu, 9 Jul 2026 16:32:45 UTC (4,075 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.