---
格式版本: 2
标题: "InceptionGS: Generative Bootstrapping for Large-Scale Gaussian Splatting under Unstructured View Sampling"
原文链接: "https://arxiv.org/abs/2609.02747"
发布日期: "2026-09-02"
发布时间校准状态: "found"
发布时间需复核: "否"
发布时间来源: "rule:local:strict_original_body"
发布时间证据: "**\\[v1\\]** Wed, 2 Sep 2026 15:48:15 UTC (6,604 KB)"
发布时间校准原因: "规则确认唯一严格发布时间，来源 local:strict_original_body"
发布时间校准置信度: "high"
发布时间候选数量: 18
发布时间严格候选数量: 6
发布时间原页读取状态: "source template page reused from URL open"
发布时间未找到原因: ""
发布时间校准时间: "2026-09-04T01:39:19+08:00"
发布时间仲裁状态: "skipped"
发布时间仲裁尝试次数: 0
发布时间仲裁耗时毫秒: 0
发现时间: "2026-09-04T01:39:03+08:00"
入库时间: "2026-09-03T17:39:19.230Z"
来源平台: "arXiv 学术论文搜索"
搜索渠道: "source_template"
搜索词: "https://arxiv.org/search/?query=Immersion&searchtype=all"
匹配关键词:
  - "Immersion"
相关厂家:
  []
相关专家:
  []
内容类型: "网页"
抓取工具: "Free Fetch + Defuddle"
清洗工具: "Defuddle Markdown + Defuddle/Readability 正文提取"
原始附件:
  []
AI优质: "否"
AI打分: 15
AI分档: "非优质"
AI质检状态: "不通过"
AI打分理由: "该资料为计算机视觉论文，研究高斯泼溅场景重建，与超节点/AI Rack、机柜级AI基础设施、供电散热互连等主题完全无关，误命中Immersion关键词，判不通过。"
AI质检模型: "zj-deepseek-v4-flash"
AI质检时间: "2026-09-04T01:39:42+08:00"
AI主题相关性: 0
AI来源权威性: 5
AI新颖性: 0
AI技术细节: 0
AI商业部署信号: 0
AI完整性: 10
AI摘要: "InceptionGS 提出一种生成式引导方法，通过平衡重建与生成来修复大规模高斯溅射中因视图采样不足导致的区域缺陷，同时保持其余区域质量。在真实大规模场景实验中，该方法提升了非结构化图像下的高保真渲染效果。"
AI摘要模型: "ali-deepseek-v4-flash"
AI摘要时间: "2026-09-04T00:17:35.505Z"
采集批次: "2026年9月3日22点43分34秒"
采集批次ID: "20260903-224334-406"
去重键: "https://arxiv.org/abs/2609.02747"
---

## Computer Science > Computer Vision and Pattern Recognition

## Title:InceptionGS: Generative Bootstrapping for Large-Scale Gaussian Splatting under Unstructured View Sampling

Authors:[Tianheng Lu](https://arxiv.org/search/cs?searchtype=author&query=Lu,+T), [Guangyu Wang](https://arxiv.org/search/cs?searchtype=author&query=Wang,+G), [Ruqi Huang](https://arxiv.org/search/cs?searchtype=author&query=Huang,+R), [Lu Fang](https://arxiv.org/search/cs?searchtype=author&query=Fang,+L)

[View PDF](https://arxiv.org/pdf/2609.02747) [HTML (experimental)](https://arxiv.org/html/2609.02747v1)

> Abstract:Achieving truly immersive large-scale scene digitization necessitates consistent and visually pleasing rendering across all possible viewing perspectives. However, collecting multi-view images covering every fine detail of a large-scale scene is prohibitive due to scene complexity, capture cost, negligence, or accessibility constraints. As a result, the sampled views tend to be highly unstructured -- the majority of the scene is well covered yet certain regions inevitably lack sufficient observations. Existing reconstruction based methods are vulnerable to view scarcity while generation based approaches suffer from generalization, controllability, and 3D consistency issues. To address this challenge, we propose InceptionGS, which bootstraps Gaussian splatting by subtly balancing reconstruction and generation. Starting from an initial Gaussian splatting, InceptionGS reasonably rethinks and repairs problematic regions caused by view scarcity while preserving the quality elsewhere, by softly incorporating scene- and view-adaptive generative priors. Extensive experiments on real-world large-scale scenes demonstrate the superiority and broad applicability of our approach in handling unstructured imagery and boosting high-fidelity Gaussian splatting. Please refer to the supplementary video for better visual demonstrations.

| Subjects: | Computer Vision and Pattern Recognition (cs.CV) |
| --- | --- |
| Cite as: | [arXiv:2609.02747](https://arxiv.org/abs/2609.02747) \[cs.CV\] |
|  | (or [arXiv:2609.02747v1](https://arxiv.org/abs/2609.02747v1) \[cs.CV\] for this version) |
|  | [https://doi.org/10.48550/arXiv.2609.02747](https://doi.org/10.48550/arXiv.2609.02747) |

## Submission history

From: Tianheng Lu \[[view email](https://arxiv.org/show-email/c3ca325e/2609.02747)\]  
**\[v1\]** Wed, 2 Sep 2026 15:48:15 UTC (6,604 KB)

[Which authors of this paper are endorsers?](https://arxiv.org/auth/show-endorsers/2609.02747) | Disable MathJax ([What is MathJax?](https://info.arxiv.org/help/mathjax.html))