WildRayZer：动态环境中的自监督大视角合成
WildRayZer: Self-supervised Large View Synthesis in Dynamic Environments
用「仅相机的静态渲染器 + 残差」做分析-合成测试，从残差自动生成伪运动掩码并蒸馏运动估计器，再以掩码 token 与门控损失梯度把监督聚焦到跨视角背景补全——这套「静态解释不了的就是瞬态」的自监督信号构造方式，可直接迁移到动态场景重建、稀疏视角 NVS 与动态 SLAM 中的运动分割。
原文依据 a camera-only static renderer explains rigid structure, and its residuals reveal transient regions
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 WildRayZer，一个面向相机与物体同时运动的动态环境的新视角合成（NVS）自监督框架。动态内容破坏了静态 NVS 模型所依赖的多视角一致性，导致重影、几何幻觉与不稳定的位姿估计。WildRayZer 通过执行分析-合成测试来解决该问题：一个仅相机的静态渲染器解释刚性结构，其残差揭示瞬态区域。基于这些残差，我们构造伪运动掩码，蒸馏出一个运动估计器，并用它掩蔽输入 token 并门控损失梯度，使监督聚焦于跨视角背景补全。为支持大规模训练与评测，我们构建了 Dynamic RealEstate10K（D-RE10K），一个包含 1.5 万条随意拍摄的动态序列的真实世界数据集，以及 D-RE10K-iPhone，一个用于稀疏视角瞬态感知 NVS 的成对瞬态与干净基准。实验表明，WildRayZer 在瞬态区域去除与全帧 NVS 质量上均以单次前馈推理稳定优于基于优化的与前馈基线。
英文原文摘要 1120 字符
We present WildRayZer, a self-supervised framework for novel view synthesis (NVS) in dynamic environments where both the camera and objects move. Dynamic content breaks the multi-view consistency that static NVS models rely on, leading to ghosting, hallucinated geometry, and unstable pose estimation. WildRayZer addresses this by performing an analysis-by-synthesis test: a camera-only static renderer explains rigid structure, and its residuals reveal transient regions. From these residuals, we construct pseudo motion masks, distill a motion estimator, and use it to mask input tokens and gate loss gradients so supervision focuses on cross-view background completion. To enable large-scale training and evaluation, we curate Dynamic RealEstate10K (D-RE10K), a real-world dataset of 15K casually captured dynamic sequences, and D-RE10K-iPhone, a paired transient and clean benchmark for sparse-view transient-aware NVS. Experiments show that WildRayZer consistently outperforms optimization-based and feed-forward baselines in both transient-region removal and full-frame NVS quality with a single feed-forward pass.