arXiv 2602.163372026-02-18cs.CVcs.LG

减法调制网络与可学习周期激活

Subtractive Modulative Network with Learnable Periodic Activations

School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Sweden（电气工程与计算机科学学院，皇家理工学院，瑞典）；School of Computer and Communication Sciences, EPFL, Switzerland（计算机与通信科学学院，瑞士联邦理工学院）

Tiou Wang, Zhuoqian Yang, Markus Flierl, Mathieu Salzmann, Sabine Süsstrunk

备注：: 4 pages, 3 figures, 3 tables
发表信息：: ICASSP 2026-2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

SMN通过可学习周期激活和调制掩码模块，实现了高效高保真图像重建和3D NeRF新视角合成。

摘要：

我们提出了一种减法调制网络（SMN），一种新颖的、参数高效的隐式神经表示（INR）架构，灵感来自经典的减法合成。SMN被设计为一个原则性的信号处理流程，具有一个可学习的周期激活层（振荡器），能够生成多频基底，并且一系列调制掩码模块（滤波器）能够主动生成高阶谐波。我们为我们的设计提供了理论分析和实证验证。我们的SMN在两个图像数据集上实现了PSNR为40+ dB，与最先进的方法相比，在重建准确性和参数效率方面都表现优异。此外，在具有挑战性的3D NeRF新视角合成任务中也观察到一致的优势。补充材料可在https://inrainbws.github.io/smn/上获得。

英文摘要：

We propose the Subtractive Modulative Network (SMN), a novel, parameter-efficient Implicit Neural Representation (INR) architecture inspired by classical subtractive synthesis. The SMN is designed as a principled signal processing pipeline, featuring a learnable periodic activation layer (Oscillator) that generates a multi-frequency basis, and a series of modulative mask modules (Filters) that actively generate high-order harmonics. We provide both theoretical analysis and empirical validation for our design. Our SMN achieves a PSNR of $40+$ dB on two image datasets, comparing favorably against state-of-the-art methods in terms of both reconstruction accuracy and parameter efficiency. Furthermore, consistent advantage is observed on the challenging 3D NeRF novel view synthesis task. Supplementary materials are available at https://inrainbws.github.io/smn/.

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。