DART: A Degradation-Aware Recurrent Transformer for Archival Film Restoration

2026-07-23

AI Summary

Historical film footage is often afflicted by compound degradations—including scratches, dust, blur, noise, flicker, and photometric aging—and lacks clean reference frames, posing significant challenges for restoration. This work proposes a degradation-aware recurrent Transformer model that introduces, for the first time in film restoration, an explicit degradation modeling mechanism. By predicting soft defect masks and propagating them temporally, the model dynamically guides temporal fusion and restoration processes. The approach markedly enhances no-reference perceptual quality, yielding cleaner and more spatiotemporally coherent results on real-world historical film data while maintaining a compact and efficient architecture.

Abstract

Archival film restoration is a challenging problem because historical footage contains compound degradations such as scratches, dust, blur, noise, flicker, and photometric aging, while clean reference videos are unavailable. Existing video restoration methods largely treat these degradations implicitly, reconstructing frames without explicit knowledge of where damage occurs or how severe it is. We propose DART, a degradation-aware recurrent transformer for archival film restoration. DART predicts and propagates a soft defect mask through time, using it to guide temporal fusion and condition the restoration network on both damage location and severity. This makes the restoration process explicitly aware of film artifacts rather than relying only on reconstruction losses. Experiments on real archival benchmarks show that DART improves no-reference perceptual quality over prior restoration architectures while remaining compact and efficient, producing cleaner and more temporally consistent restorations of structured film damage.

Problem

• archival film restoration
• compound degradations
• defect awareness
• temporal consistency
• no-reference restoration

Innovation

• degradation-aware
• soft defect mask
• recurrent transformer
• archival film restoration
• temporal fusion