SNAP：从新视角合成学习几何表征的自监督Transformer
Less Decoder is More Encoder: Geometric Representation Learning from Novel View Synthesis
一篇研究 NVS 架构选择的论文：SNAP 只靠限制 decoder 表达力加 latent 重建，五个几何任务上就追平了专门监督方法，做 3D 视觉的可以看看。
论文提出 SNAP，一个自监督 encoder-decoder transformer，用于从 Novel View Synthesis 中学习几何表征。它用 pose-conditioned local decoder 和 latent-space 重建目标，解决了空间表达力过强的 decoder 稀释 scene encoder 能力、像素级目标阻碍特征学习这两个问题。在视觉定位、位姿估计、点对应、深度估计、机器人操作五个任务上，SNAP 与专门的几何监督方法及自监督表征相比均有竞争力。其 patch 特征在更低算力和数据预算下涌现出接近重度监督模型的视角不变性，在相机偏移导致标准 2D 表征失效的场景中退化更平缓。
Less Decoder is More Encoder: Geometric Representation Learning from Novel View Synthesis
This paper examines the role of Novel View Synthesis (NVS) in geometric representation learning. In principle, NVS should reason about 3D scene structure, thereby enabling transferable multi-view geometric representations. Yet, existing encoder-based NVS methods yield poor representations. This is not because of a lack of supervisory signal, but rather due to inconspicuous architectural choices: \textit{spatially expressive decoders} that dilute representational capabilities of the scene encoder, and \textit{low-level pixel-space targets} that hinder feature learning. We present SNAP, a self-supervised encoder-decoder transformer that addresses both through a pose-conditioned local decoder and a latent-space reconstruction objective. SNAP is task agnostic, and we show that it is competitive with special-purpose geometry-supervised methods. SNAP also performs competitively against self-supervised representations across five tasks: visual localization, pose estimation, point correspondence, depth estimation, and robot manipulation. Remarkably, SNAP's patch features exhibit emergent viewpoint invariance that approaches heavily supervised models despite lower compute and data budgets. Under camera shifts where standard 2D representations collapse, SNAP degrades more gracefully, revealing that restricting decoder expressivity actively prevents the suppression of transferable geometric structure. https://snap-nvs.github.io
