Moiré artifacts are common in digital photography, resulting from the interference between high-frequency scene content and the color filter array of the camera. Existing deep learning-based demoiréing methods trained on large scale datasets are limited in handling various complex moiré patterns, and mainly focus on demoiréing of photos taken of digital displays. Moreover, obtaining moiré-free ground-truth in natural scenes is difficult but needed for training. In this paper, we propose a self-adaptive learning method for demoiréing a high-frequency image, with the help of an additional defocused moiré-free blur image. Given an image degraded with moiré artifacts and a moiré-free blur image, our network predicts a moiré-free clean image and a blur kernel with a self-adaptive strategy that does not require an explicit training stage, instead performing test-time adaptation.

We use three cameras and three screens to capture our dataset; please see Table 1 for the specifications. The following algorithms (Procedure 1 and Procedure 2) show the joint optimization method and the baseline alternating optimization method compared in the ablation study in Section 5.1 of the main paper. Procedure 1 The joint optimization algorithm.Input: Focused image M with moiré patterns and defocused blur image B without moiré patterns; Output: Estimated moiré-free image C; Procedure 2 The alternating optimization algorithm.Input: Focused image M with moiré patterns and defocused blur image B without moiré patterns; Output: Estimated moiré-free image C; We also test our model on a smartphone HUA WEI P30 PRO. To test on the real world examples, we do some preprocessing, e.g., alignment. Figure 1: Examples captured from natural scenes.

Moiré artifacts are common in digital photography, resulting from the interference between high-frequency scene content and the color filter array of the camera. Existing deep learning-based demoiréing methods trained on large scale datasets are limited in handling various complex moiré patterns, and mainly focus on demoiréing of photos taken of digital displays. Moreover, obtaining moiré-free ground-truth in natural scenes is difficult but needed for training. In this paper, we propose a self-adaptive learning method for demoiréing a high-frequency image, with the help of an additional defocused moiré-free blur image. Given an image degraded with moiré artifacts and a moiré-free blur image, our network predicts a moiré-free clean image and a blur kernel with a self-adaptive strategy that does not require an explicit training stage, instead performing test-time adaptation.