SceneTok：面向三维场景的压缩可扩散 token 空间
SceneTok: A Compressed, Diffusable Token Space for 3D Scenes
把多视角图像编码成一小撮置换不变、脱离空间网格的潜变量 token，再用轻量整流流解码器渲染新视角，为 3D 场景生成提供了比 3DGS/NeRF 更紧凑、可直接接扩散/流模型的表示接口。
原文依据 a novel tokenizer for encoding view sets of scenes into a compressed and diffusable set of unstructured tokens
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 SceneTok，一种新颖的分词器，用于将场景的视图集合编码为压缩且可扩散的无结构 token 集合。现有的三维场景表示与生成方法通常使用三维数据结构或视图对齐的场。与之不同，我们首次提出将场景信息编码为一小组置换不变 token 的方法，且该表示与空间网格解耦。场景 token 由多视图分词器在给定大量上下文视图时预测，并通过轻量级整流流解码器渲染为新视图。我们表明，该表示的压缩率比其他表示强 1 至 3 个数量级，同时仍达到最优的重建质量。此外，该表示可从新的轨迹渲染，包括偏离输入轨迹的轨迹，并且我们表明解码器能够优雅地处理不确定性。最后，高度压缩的无结构潜变量场景 token 集合使得在 5 秒内完成简单高效场景生成成为可能，实现了比以往范式更好的质量与速度权衡。
英文原文摘要 1126 字符
We present SceneTok, a novel tokenizer for encoding view sets of scenes into a compressed and diffusable set of unstructured tokens. Existing approaches for 3D scene representation and generation commonly use 3D data structures or view-aligned fields. In contrast, we introduce the first method that encodes scene information into a small set of permutation-invariant tokens that is disentangled from the spatial grid. The scene tokens are predicted by a multi-view tokenizer given many context views and rendered into novel views by employing a light-weight rectified flow decoder. We show that the compression is 1-3 orders of magnitude stronger than for other representations while still reaching state-of-the-art reconstruction quality. Further, our representation can be rendered from novel trajectories, including ones deviating from the input trajectory, and we show that the decoder gracefully handles uncertainty. Finally, the highly-compressed set of unstructured latent scene tokens enables simple and efficient scene generation in 5 seconds, achieving a much better quality-speed trade-off than previous paradigms.