[2512.05106] NeuralRemaster: Phase-Preserving Diffusion for Structure-Aligned Generation

Computer Science > Computer Vision and Pattern Recognition arXiv:2512.05106 (cs) [Submitted on 4 Dec 2025 (v1), last revised 5 Mar 2026 (this version, v3)] Title:NeuralRemaster: Phase-Preserving Diffusion for Structure-Aligned Generation Authors:Yu Zeng, Charles Ochoa, Mingyuan Zhou, Vishal M. Patel, Vitor Guizilini, Rowan McAllister View a PDF of the paper titled NeuralRemaster: Phase-Preserving Diffusion for Structure-Aligned Generation, by Yu Zeng and 5 other authors View PDF HTML (experimental) Abstract:Standard diffusion corrupts data using Gaussian noise whose Fourier coefficients have random magnitudes and random phases. While effective for unconditional or text-to-image generation, corrupting phase components destroys spatial structure, making it ill-suited for tasks requiring geometric consistency, such as re-rendering, simulation enhancement, and image-to-image translation. We introduce Phase-Preserving Diffusion (\phi-PD), a model-agnostic reformulation of the diffusion process that preserves input phase while randomizing magnitude, enabling structure-aligned generation without architectural changes or additional parameters. We further propose Frequency-Selective Structured (FSS) noise, which provides continuous control over structural rigidity via a single frequency-cutoff parameter. \phi-PD adds no inference-time cost and is compatible with any diffusion model for images or videos. Across photorealistic and stylized re-rendering, as well as sim-to-real enhance...