arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.11141cs.HCcs.LG
收藏

LCIP: 高维图像数据的损失控制逆投影

LCIP: Loss-Controlled Inverse Projection of High-Dimensional Image Data

Utrecht University(乌特雷赫大学);University of Konstanz(康斯坦茨大学)

Yu Wang, Frederik L. Dennig, Michael Behrisch, Alexandru Telea

AI总结:

LCIP通过用户控制的扫描机制实现高维图像数据的逆投影,提升数据空间的覆盖能力,适用于图像风格迁移等任务。

摘要:

投影(或降维)方法P旨在将高维数据映射到通常为2D散点图用于可视化探索。逆投影方法P⁻¹旨在将此2D空间映射到数据空间以支持数据增强、分类器分析和数据插补等任务。当前P⁻¹方法存在根本性限制——它们只能生成数据空间中的固定表面状结构,这无法充分覆盖该空间的丰富性。我们通过一种新的方法来解决这一问题,该方法可以由用户控制来“扫描”数据空间。我们的方法适用于任何P技术及数据集,由两个直观的用户设置参数控制,并且易于实现。我们通过广泛的图像处理应用来展示,包括风格迁移。

英文摘要:

Projections (or dimensionality reduction) methods $P$ aim to map high-dimensional data to typically 2D scatterplots for visual exploration. Inverse projection methods $P^{-1}$ aim to map this 2D space to the data space to support tasks such as data augmentation, classifier analysis, and data imputation. Current $P^{-1}$ methods suffer from a fundamental limitation -- they can only generate a fixed surface-like structure in data space, which poorly covers the richness of this space. We address this by a new method that can `sweep' the data space under user control. Our method works generically for any $P$ technique and dataset, is controlled by two intuitive user-set parameters, and is simple to implement. We demonstrate it by an extensive application involving image manipulation for style transfer.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑