pipette
ENEnglish

Dual-Scale Condition-Shift Editing for Robust Real-World Image Demoir\'eing

Shiyang Zhou, Haijin Zeng, Congpei Qiu, Zhengcen Li, Ping Chen, Xiangming Wang, Zheng Li, Yongyong Chen, Jingyong Su

PreprintUso en el mundo real

En palabras de los autores

Robust demoir\'eing of screen-recaptured images remains challenging: moir\'e varies in frequency, color, and spatial support across display-camera pairs and capture conditions. Dedicated restoration networks trained on finite paired data struggle with this diversity, especially when low-frequency color banding and high-frequency stripes co-occur. We present Demoir\'eEdit, a content-preserving editing framework that adapts the natural-image prior of a pretrained flow-matching editor to demoir\'eing. First, an endpoint-supervised training objective supplements velocity matching with pixel, perceptual, and frequency-domain restoration losses. Second, a dual-scale condition-shift strategy performs global correction at low resolution, then refines the image at full resolution by shifting the denoising condition from the first-stage output to the original input for detail recovery. We also introduce DailyMoir\'e130, a benchmark of challenging moir\'e in everyday screen photography, comprising 130 smartphone captures across diverse displays with low-frequency color banding, fine high-frequency stripes, and mixed-frequency interference. On the challenging DailyMoir\'e130 benchmark, our method achieves the lowest residual moir\'e area and receives 71.3% overall-quality preference in a blind Best-of-7 study with 31 participants. We also evaluate on paired FHDMi/UHDM, reducing residual moir\'e confidence by 47.1%/9.3% and increasing MUSIQ by 4.3%/2.0% over the best baseline for each metric.

Resultado principalEl resumen no menciona limitaciones.

Apareció: martes, 20 de octubre. arXiv. Preprint, todavía sin revisión por pares.