从无到有：通过新视角合成的自监督三维重建
From None to All: Self-Supervised 3D Reconstruction via Novel View Synthesis
把相机参数预测与显式 3D 高斯重建放进同一个共享 Transformer 骨干网络、用掩码注意力约束，仅靠 2D 光度监督即可从未标定、无位姿的图像集合前馈重建，为无标注数据上的可扩展三维重建提供了可复用的自监督训练范式。
原文依据 reconstructs 3D Gaussians from uncalibrated and unposed context views and renders target views using its self-predicted camera parameters
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论本文提出 NAS3R，一个自监督前馈框架，在没有任何真值标注、也没有预训练先验的条件下联合学习显式三维几何与相机参数。训练过程中，NAS3R 从未标定、无位姿的上下文视角重建三维高斯，并利用其自预测的相机参数渲染目标视角，从而在二维光度监督下实现自监督训练。为保证收敛稳定，NAS3R 将重建与相机预测集成在由掩码注意力调控的共享 Transformer 骨干网络之中，并采用基于深度的高斯表示以促成良态的优化。该框架与最优的监督式三维重建架构兼容，并可在可用时引入预训练先验或内参信息。大量实验表明，NAS3R 取得了优于其他自监督方法的结果，为从无约束数据中进行三维重建建立了一种可扩展且具备几何感知的范式。代码与模型已公开于 https://ranrhuang.github.io/nas3r/。
英文原文摘要 1119 字符
In this paper, we introduce NAS3R, a self-supervised feed-forward framework that jointly learns explicit 3D geometry and camera parameters with no ground-truth annotations and no pretrained priors. During training, NAS3R reconstructs 3D Gaussians from uncalibrated and unposed context views and renders target views using its self-predicted camera parameters, enabling self-supervised training from 2D photometric supervision. To ensure stable convergence, NAS3R integrates reconstruction and camera prediction within a shared transformer backbone regulated by masked attention, and adopts a depth-based Gaussian formulation that facilitates well-conditioned optimization. The framework is compatible with state-of-the-art supervised 3D reconstruction architectures and can incorporate pretrained priors or intrinsic information when available. Extensive experiments show that NAS3R achieves superior results to other self-supervised methods, establishing a scalable and geometry-aware paradigm for 3D reconstruction from unconstrained data. Code and models are publicly available at https://ranrhuang.github.io/nas3r/.