Newest papers

Dual-Scale Condition-Shift Editing for Robust Real-World Image Demoiréing

Shiyang Zhou, Haijin Zeng, Congpei Qiu, Zhengcen Li, Ping Chen, Xiangming Wang, Zheng Li, Yongyong Chen, Jingyong Su · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.26644 · cited by 0 works (OpenAlex) · open-access PDF

Indonesian

Rangkaian pemulihan yang kuat dari gambar-gambar yang direkam layar tetap menantang: moiré bervariasi dalam frekuensi, warna, dan dukungan spasial di antara pasangan kamera display dan kondisi penangkapan. Jaringan pemulihan yang didedikasikan dilatih dalam pertarungan data berpasangan finite dengan keragaman ini, terutama ketika banding warna frekuensi rendah dan strip frekuensi tinggi terjadi bersama-sama. Kami menyajikan DemoiréEdit, kerangka kerja pengeditan berbasis konten yang menyesuaikan gambar-gambar alami sebelum pengeditan flow-matching pra-latihan untuk demoiré. Pertama, objektif pelatihan yang diawasi oleh titik akhir menambahkan kecepatan yang cocok dengan piksel, perceptual, dan kehilangan pemulihan domain frekuensi tinggi. Kedua, strategi perubahan kondisi dua skala Mi melakukan koreksi global

Machine translation (Meta M2M100); the English original follows.

Abstract

Robust demoiréing of screen-recaptured images remains challenging: moiré 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 DemoiréEdit, a content-preserving editing framework that adapts the natural-image prior of a pretrained flow-matching editor to demoiréing. 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 DailyMoiré130, a benchmark of challenging moiré 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 DailyMoiré130 benchmark, our method achieves the lowest residual moiré 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 moiré confidence by 47.1%/9.3% and increasing MUSIQ by 4.3%/2.0% over the best baseline for each metric.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Artificial intelligence, Computer science, Residual, Computer vision, Benchmark (surveying), Image editing

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange