SEM-Net: Efficient Pixel Modelling for image inpainting with Spatially Enhanced SSM

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Shuang, Zhang, Haozheng, Atapour-Abarghouei, Amir, Shum, Hubert P. H.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929586289246208
author Chen, Shuang
Zhang, Haozheng
Atapour-Abarghouei, Amir
Shum, Hubert P. H.
author_facet Chen, Shuang
Zhang, Haozheng
Atapour-Abarghouei, Amir
Shum, Hubert P. H.
contents Image inpainting aims to repair a partially damaged image based on the information from known regions of the images. \revise{Achieving semantically plausible inpainting results is particularly challenging because it requires the reconstructed regions to exhibit similar patterns to the semanticly consistent regions}. This requires a model with a strong capacity to capture long-range dependencies. Existing models struggle in this regard due to the slow growth of receptive field for Convolutional Neural Networks (CNNs) based methods and patch-level interactions in Transformer-based methods, which are ineffective for capturing long-range dependencies. Motivated by this, we propose SEM-Net, a novel visual State Space model (SSM) vision network, modelling corrupted images at the pixel level while capturing long-range dependencies (LRDs) in state space, achieving a linear computational complexity. To address the inherent lack of spatial awareness in SSM, we introduce the Snake Mamba Block (SMB) and Spatially-Enhanced Feedforward Network. These innovations enable SEM-Net to outperform state-of-the-art inpainting methods on two distinct datasets, showing significant improvements in capturing LRDs and enhancement in spatial consistency. Additionally, SEM-Net achieves state-of-the-art performance on motion deblurring, demonstrating its generalizability. Our source code will be released in https://github.com/ChrisChen1023/SEM-Net.
format Preprint
id arxiv_https___arxiv_org_abs_2411_06318
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SEM-Net: Efficient Pixel Modelling for image inpainting with Spatially Enhanced SSM
Chen, Shuang
Zhang, Haozheng
Atapour-Abarghouei, Amir
Shum, Hubert P. H.
Computer Vision and Pattern Recognition
Image inpainting aims to repair a partially damaged image based on the information from known regions of the images. \revise{Achieving semantically plausible inpainting results is particularly challenging because it requires the reconstructed regions to exhibit similar patterns to the semanticly consistent regions}. This requires a model with a strong capacity to capture long-range dependencies. Existing models struggle in this regard due to the slow growth of receptive field for Convolutional Neural Networks (CNNs) based methods and patch-level interactions in Transformer-based methods, which are ineffective for capturing long-range dependencies. Motivated by this, we propose SEM-Net, a novel visual State Space model (SSM) vision network, modelling corrupted images at the pixel level while capturing long-range dependencies (LRDs) in state space, achieving a linear computational complexity. To address the inherent lack of spatial awareness in SSM, we introduce the Snake Mamba Block (SMB) and Spatially-Enhanced Feedforward Network. These innovations enable SEM-Net to outperform state-of-the-art inpainting methods on two distinct datasets, showing significant improvements in capturing LRDs and enhancement in spatial consistency. Additionally, SEM-Net achieves state-of-the-art performance on motion deblurring, demonstrating its generalizability. Our source code will be released in https://github.com/ChrisChen1023/SEM-Net.
title SEM-Net: Efficient Pixel Modelling for image inpainting with Spatially Enhanced SSM
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.06318