HaineiFRDM: Structure-Preserving Diffusion for Film Restoration under Fast Motion and Diverse Defects

[Submitted on 31 Dec 2025 (v1), last revised 24 Jun 2026 (this version, v3)]

View PDF HTML (experimental)

Abstract:Existing film-restoration methods frequently fail under fast motion, producing limb disappearance and structural distortion due to inaccurate motion modeling. Moreover, high-resolution restoration under spatially-persistent and mixed defects remains insufficiently studied. We propose HaineiFRDM, a Film Restoration Diffusion Model that leverages the content modeling capability of diffusion models for content-aware restoration, removing defects while preserving scene this http URL enable scalable high-resolution restoration, we adopt a patch-wise strategy with position-aware global fusion modules to maintain cross-patch coherence. We further introduce a frequency-based module to enhance texture consistency and a patch-consistent inference framework to alleviate blocking artifacts introduced by patch-based this http URL also construct a film restoration dataset comprising categorized defect templates, professionally restored films, and realistic synthetic this http URL experiments demonstrate our superior restoration quality with strong structural consistency. Our design also reduces memory requirements, enabling high-resolution restoration on a single 24GB-VRAM this http URL and the dataset will be released at this https URL.

Submission history

From: Rongji Xun [view email]
[v1] Wed, 31 Dec 2025 16:18:07 UTC (39,579 KB)
[v2] Sun, 21 Jun 2026 17:33:59 UTC (40,493 KB)
[v3] Wed, 24 Jun 2026 14:47:42 UTC (40,493 KB)