从射线到投影：为前馈视图合成提供更好的输入
From Rays to Projections: Better Inputs for Feed-Forward View Synthesis
用目标视图的投影线索替代 Plücker 射线图作为相机条件输入，把脆弱的射线空间几何回归转化为良态的目标视图图像到图像翻译问题；做前馈新视角合成时可直接替换相机编码方式，并配套设计了适配该线索的掩码自编码预训练以利用无标定数据。
原文依据 We propose projective conditioning, which replaces raw camera parameters with a target-view projective cue that provides a stable 2D input.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论前馈视图合成模型以极小的三维归纳偏置在单次前向中预测新视角。现有工作将相机编码为 Plücker 射线图，这使预测与任意的世界坐标基准绑定，并对微小的相机变换敏感，从而削弱了几何一致性。本文追问：什么样的输入最能为一模型提供条件，使其实现鲁棒且一致的视图合成。我们提出投影条件化，用目标视图的投影线索替代原始相机参数，从而提供稳定的二维输入。这将任务从射线空间中脆弱的几何回归问题，重新表述为良态的目标视图图像到图像翻译问题。此外，我们引入一种针对该线索定制的掩码自编码预训练策略，使得能够利用大规模未标定数据进行预训练。在我们的视图一致性基准上，与射线条件化的基线相比，本方法展现出更高的保真度和更强的跨视图一致性；在标准新视角合成基准上也取得了最优的质量。
英文原文摘要 1096 字符
Feed-forward view synthesis models predict a novel view in a single pass with minimal 3D inductive bias. Existing works encode cameras as Plucker ray maps, which tie predictions to the arbitrary world coordinate gauge and make them sensitive to small camera transformations, thereby undermining geometric consistency. In this paper, we ask what inputs best condition a model for robust and consistent view synthesis. We propose projective conditioning, which replaces raw camera parameters with a target-view projective cue that provides a stable 2D input. This reframes the task from a brittle geometric regression problem in ray space to a well-conditioned target-view image-to-image translation problem. Additionally, we introduce a masked autoencoding pretraining strategy tailored to this cue, enabling the use of large-scale uncalibrated data for pretraining. Our method shows improved fidelity and stronger cross-view consistency compared to ray-conditioned baselines on our view-consistency benchmark. It also achieves state-of-the-art quality on standard novel view synthesis benchmarks.