Efficient Moiré pattern removal with lightweight multi-scale feature extraction

In recent years, convolutional neural networks have excelled in image Moir & eacute; pattern removal, yet their high memory consumption poses challenges for resource-constrained devices. To address this, we propose the lightweight multi-scale network (LMSNet). Designing lightweight multi-scale feature extraction blocks and efficient adaptive channel fusion modules, we extend the receptive field of feature extraction and introduce lightweight convolutional decomposition. LMSNet achieves a balance between parameter numbers and reconstruction performance. Extensive experiments demonstrate that our LMSNet, with 0.77 million parameters, achieves Moir & eacute; pattern removal performance comparable to full high definition demoir & eacute;ing network (FHDe(2)Net) with 13.57 million parameters.