Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.03077 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908479161106432 |
|---|---|
| author | Wu, Anran Peng, Long Di, Xin Dai, Xueyuan Wu, Chen Wang, Yang Fu, Xueyang Cao, Yang Zha, Zheng-Jun |
| author_facet | Wu, Anran Peng, Long Di, Xin Dai, Xueyuan Wu, Chen Wang, Yang Fu, Xueyang Cao, Yang Zha, Zheng-Jun |
| contents | Feedforward 3D Gaussian Splatting (3DGS) overcomes the limitations of optimization-based 3DGS by enabling fast and high-quality reconstruction without the need for per-scene optimization. However, existing feedforward approaches typically assume that input multi-view images are clean and high-quality. In real-world scenarios, images are often captured under challenging conditions such as noise, low light, or rain, resulting in inaccurate geometry and degraded 3D reconstruction. To address these challenges, we propose a general and efficient multi-view feature enhancement module, RobustGS, which substantially improves the robustness of feedforward 3DGS methods under various adverse imaging conditions, enabling high-quality 3D reconstruction. The RobustGS module can be seamlessly integrated into existing pretrained pipelines in a plug-and-play manner to enhance reconstruction robustness. Specifically, we introduce a novel component, Generalized Degradation Learner, designed to extract generic representations and distributions of multiple degradations from multi-view inputs, thereby enhancing degradation-awareness and improving the overall quality of 3D reconstruction. In addition, we propose a novel semantic-aware state-space model. It first leverages the extracted degradation representations to enhance corrupted inputs in the feature space. Then, it employs a semantic-aware strategy to aggregate semantically similar information across different views, enabling the extraction of fine-grained cross-view correspondences and further improving the quality of 3D representations. Extensive experiments demonstrate that our approach, when integrated into existing methods in a plug-and-play manner, consistently achieves state-of-the-art reconstruction quality across various types of degradations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_03077 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RobustGS: Unified Boosting of Feedforward 3D Gaussian Splatting under Low-Quality Conditions Wu, Anran Peng, Long Di, Xin Dai, Xueyuan Wu, Chen Wang, Yang Fu, Xueyang Cao, Yang Zha, Zheng-Jun Computer Vision and Pattern Recognition Feedforward 3D Gaussian Splatting (3DGS) overcomes the limitations of optimization-based 3DGS by enabling fast and high-quality reconstruction without the need for per-scene optimization. However, existing feedforward approaches typically assume that input multi-view images are clean and high-quality. In real-world scenarios, images are often captured under challenging conditions such as noise, low light, or rain, resulting in inaccurate geometry and degraded 3D reconstruction. To address these challenges, we propose a general and efficient multi-view feature enhancement module, RobustGS, which substantially improves the robustness of feedforward 3DGS methods under various adverse imaging conditions, enabling high-quality 3D reconstruction. The RobustGS module can be seamlessly integrated into existing pretrained pipelines in a plug-and-play manner to enhance reconstruction robustness. Specifically, we introduce a novel component, Generalized Degradation Learner, designed to extract generic representations and distributions of multiple degradations from multi-view inputs, thereby enhancing degradation-awareness and improving the overall quality of 3D reconstruction. In addition, we propose a novel semantic-aware state-space model. It first leverages the extracted degradation representations to enhance corrupted inputs in the feature space. Then, it employs a semantic-aware strategy to aggregate semantically similar information across different views, enabling the extraction of fine-grained cross-view correspondences and further improving the quality of 3D representations. Extensive experiments demonstrate that our approach, when integrated into existing methods in a plug-and-play manner, consistently achieves state-of-the-art reconstruction quality across various types of degradations. |
| title | RobustGS: Unified Boosting of Feedforward 3D Gaussian Splatting under Low-Quality Conditions |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.03077 |