发表机构
University College London; Meta Reality Labs; Karlsruhe Institute of Technology(伦敦大学学院; Meta现实实验室; 卡尔斯鲁厄理工学院)
机构由 AI 辅助整理,请以论文原文为准。AI 中文总结
本研究提出一种结合“草图绘制”概念的联合蒙特卡洛估计方法,生成受控的时变高斯噪声,用于下游任务的时间控制与一致性,其代码更简单、运行速度更快。
AI 中文摘要
我们提出一种生成具有受控方差和受控时间相关性的时变高斯噪声的方法,该噪声可用于多个下游任务以实现时间控制和时间一致性。核心技术思路是将该问题表述为联合蒙特卡洛估计,即结合数据库文献中的“草图绘制(sketching)”概念,同时进行经典像素重建和方差估计。我们证明,与现有方法相比,该方法能为下游任务提供时间控制,且代码更简单、运行速度更快。
英文摘要
We suggest a method to generate time-varying Gaussian noise with controlled variance and controlled temporal correlation. This noise is used in several downstream tasks for temporal control and temporal coherence. The core technical idea is to phrase this problem as joint Monte-Carlo estimation of both a classic pixel reconstruction and estimation of variance using the concept of "sketching" from the database literature. We demonstrate that our method allows temporal control for downstream tasks with simpler and faster code than previous methods.
CommentsSIGGRAPH Asia 2026 Conference Papers. Code: https://github.com/facebookresearch/Tabula-Rasa