StructSplat: 从无标定稀疏视图进行可泛化的3D高斯泼溅
StructSplat: Generalizable 3D Gaussian Splatting from Uncalibrated Sparse Views
AI总结：
提出StructSplat，一种前馈式可泛化3D高斯重建框架，无需相机参数，通过结构化表示分离几何、语义和纹理线索，在多个基准上显著超越现有方法。
摘要：
我们提出了StructSplat，一种前馈式且可泛化的3D高斯重建框架，它直接对未标定图像进行操作，无需相机参数。现有方法要么依赖于逐场景优化，要么假设已知的相机位姿，并且通常将几何和外观纠缠在一个统一的骨干网络中，限制了重建保真度和泛化能力。我们的关键思想是采用一种结构化表示，在重建过程中将几何、语义和纹理线索组织起来并赋予明确的角色。具体来说，我们引入了一种像素对齐的特征注入机制，以实现从2D观测中精确的纹理建模；融入语义感知先验以提高全局一致性；并设计了一种相机对齐策略以防止信息泄露并提高泛化能力。实验表明，我们的方法在具有挑战性的基准上显著优于先前的方法。在DL3DV上，我们的方法达到了28.045 PSNR，比AnySplat（22.377）高出+5.67 dB。在跨数据集评估中，我们的方法在ACID上比AnySplat高出+1.94 dB，在RealEstate10K上高出+1.72 dB。项目页面：此 https URL 代码：此 https URL
英文摘要：
We present StructSplat, a feed-forward and generalizable 3D Gaussian reconstruction framework that operates directly on uncalibrated images without requiring camera parameters. Existing methods either rely on per-scene optimization or assume known camera poses, and often entangle geometry and appearance within a unified backbone, limiting reconstruction fidelity and generalization. Our key idea is to adopt a structured representation that organizes geometry, semantic, and texture cues with explicit roles in the reconstruction process. Specifically, we introduce a pixel-aligned feature injection mechanism to enable accurate texture modeling from 2D observations, incorporate semantic-aware priors to improve global consistency, and design a camera alignment strategy to prevent information leakage and improve generalization. Experiments show that our method significantly outperforms prior approaches on challenging benchmarks. On DL3DV, our method achieves 28.045 PSNR, surpassing AnySplat (22.377) by +5.67 dB. In cross-dataset evaluation, our method achieves +1.94 dB over AnySplat on ACID and +1.72 dB on RealEstate10K. Project page: https://structsplat.github.io Code: https://github.com/J-C-Zhao/StructSplat
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
