[论文] Micro-Edits: Computational Support for Fine-Scale Edits in Film

论文概要

研究领域: CV
作者: Nhan Tran, Hijung Valentina Shin, Abe Davis, Mackenzie Leake
发布时间: 2026-10-29
arXiv: 2610.40653

中文摘要

一个估计器可以在典型数据上准确且稳定，但对少数可能的记录敏感。差分隐私机制通常将噪声缩放到单条记录的最坏情况影响，这让这些记录即使在其余数据允许更小噪声的情况下仍主导噪声水平。如果先移除它们就能恢复更小的噪声尺度，但移除操作本身必须是隐私的。我们提出了 propose-filter-release（PFR）框架，通过以隐私方式过滤其输入，将一个保护添加或移除典型记录（即适合剩余数据的记录）的机制，转变为在每种输入上都满足标准差分隐私的机制。利用我们新颖的过滤步骤，PFR仅需要一个凹的友好度分数（衡量一条记录与数据集的适配程度）和一个友好-DP的释放机制（即当被更改的记录适合剩余数据时满足隐私）。对于每个这样的分数和释放机制，PFR在关于记录数量的公共条件下保证差分隐私，因此它是一个通用的隐私原语，可与 propose-test-release 相媲美，FriendlyCore 可以视为其一种实现。我们通过将其实例化到标量统计量、主子空间估计、用户直方图的KL散度、图的有效电阻和处理规则选择上来展示PFR的通用性。

原文摘要

In film editing, every frame matters. While macro, content-level edits shape the high-level narrative of a film, small local edits often control timing and direct visual attention. For example, editors can adjust the precise timing of clips to control the length of a meaningful silence, or cut to shots that highlight a specific character's reactions to important moments. These small, detailed edits are slow and tedious to make in standard editing tools because users have to map high-level narrative goals to low-level frame-by-frame adjustments. At the same time, these fine-scale adjustments, which we call micro-edits, can be nuanced and expressive, with different choices serving different creative goals. In this work, we start by conducting a formative study with professional editors to un...

自动采集于 2026-10-31

#论文 #arXiv #CV #小凯