无需高质量初始点云，用结构注意力实现3D高斯溅射的直接重建
AttentionGS: Towards Initialization-Free 3D Gaussian Splatting via Structural Attention
- 通过几何与纹理双注意力机制，从随机初始化直接重建3D场景
- 在低纹理或视角受限场景中性能显著优于现有方法
- 适合真实复杂环境下的鲁棒3D重建，无需依赖SfM预处理
3D高斯溅射（3DGS）是神经辐射场（NeRF）的强大替代方案，在复杂场景重建和高效渲染方面表现优异。然而，其依赖于结构光测（SfM）生成的高质量点云，限制了应用范围。当场景缺乏纹理或视角受限时，SfM失效，导致3DGS重建严重退化。为此，我们提出AttentionGS，一种新框架，通过结构注意力实现从随机初始化直接进行3D重建，摆脱对高质量初始点云的依赖。训练初期引入几何注意力以快速恢复全局场景结构；随着训练推进，加入纹理注意力以细化细节并提升渲染质量。此外，采用透明度加权梯度引导高斯点密度增长，改善表面重建效果。在多个基准数据集上的大量实验表明，AttentionGS在点云初始化不可靠的情况下显著优于当前最优方法。该方法为真实应用中的更鲁棒、灵活的3DGS铺平道路。
原文摘要 · Abstract (English)
3D Gaussian Splatting (3DGS) is a powerful alternative to Neural Radiance Fields (NeRF), excelling in complex scene reconstruction and efficient rendering. However, it relies on high-quality point clouds from Structure-from-Motion (SfM), limiting its applicability. SfM also fails in texture-deficient or constrained-view scenarios, causing severe degradation in 3DGS reconstruction. To address this limitation, we propose AttentionGS, a novel framework that eliminates the dependency on high-quality initial point clouds by leveraging structural attention for direct 3D reconstruction from randomly initialization. In the early training stage, we introduce geometric attention to rapidly recover the global scene structure. As training progresses, we incorporate texture attention to refine fine-grained details and enhance rendering quality. Furthermore, we employ opacity-weighted gradients to guide Gaussian densification, leading to improved surface reconstruction. Extensive experiments on multiple benchmark datasets demonstrate that AttentionGS significantly outperforms state-of-the-art methods, particularly in scenarios where point cloud initialization is unreliable. Our approach paves the way for more robust and flexible 3D Gaussian Splatting in real-world applications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。