Single image super-resolution reconstruction based on multi-scale feature mapping adversarial network

Zhou, Dengwen; Duan, Ran; Zhao, Lijuan; Chai, Xiao-liang

doi:10.1016/j.sigpro.2019.107251

Cited by 20 publications

(12 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Generative Adversarial Networks (GAN) have been found to have an essential role in SR research [19]- [23]. Ledig et al [24], focus on single image SR (SISR) and present a GAN for super image resolution (SRGAN).…”

Section: B Super Resolutionmentioning

confidence: 99%

See 1 more Smart Citation

Merging Super Resolution and Attribute Learning for Low-Resolution Person Attribute Recognition

Abbaszadi

Ikizler-Cinbis

2022

IEEE Access

View full text Add to dashboard Cite

In video surveillance, visual person attributes such as gender, backpack, type of clothing are crucial for a person searching or re-identification. For detecting and retrieving these attributes with high accuracy, the availability of high-quality videos is a necessity in general. However, this cannot be guaranteed in general surveillance videos or images; beside improving hardware technology, improving inference algorithms on low-resolution data becomes valuable. The present paper attempts to address two main problems: designing neural network architecture from existing proven architectures, and a novel approach toward re-identification on low-resolution videos. The proposed architecture takes the advantage of Super-Resolution and Multi-Attribute Recognition models for more effective recognition (SRMAR). Experiments on two benchmark datasets demonstrating the effectiveness and applicability of the proposed neural network architecture for low-resolution multiple attribute recognition. Furthermore, a higher-level linear combination scheme that optimally combines the proposed architecture (SRMAR) and Multi-Attribute recognition network outperforms the two as illustrated in the experimental results.

show abstract

Section: B Super Resolutionmentioning

confidence: 99%

“…Market-1501 [9] is one of the largest person re-ID datasets, containing 32,668 images and 3,368 query images. 751 identities (19,732) are used for training and 750 identities (13,328 images) are for testing [8]. There are 27 attributes 1 provided in the dataset.…”

Section: Experiments a Datasetsmentioning

confidence: 99%

Merging Super Resolution and Attribute Learning for Low-Resolution Person Attribute Recognition

Abbaszadi

Ikizler-Cinbis

2022

IEEE Access

View full text Add to dashboard Cite

show abstract

“…However, the features from convolutional learning will gradually become abstract as the network became deeper. Many meaningful multiscale detail features may be lost [34,47,48,49], which should be considered in the sketch extraction.…”

Section: Fine Extraction Based On Msu-netmentioning

confidence: 99%

A relic sketch extraction framework based on detail-aware hierarchical deep network

Peng

Wang

et al. 2021

Signal Processing

View full text Add to dashboard Cite

“…This problem exists in both VI and IR image SR areas. Recently, in the VI image SR area, numerous methods [5][6][7][8][9][10][11][12][13][14][15][16][17][18] have been developed to find the most likely solutions. The existing VI SR methods can be roughly grouped into three categories: interpolation-based [5], prior-based [6], and learningbased methods [9][10][11][12][13][14][15][16][17][18][19][20].…”

Section: Introductionmentioning

confidence: 99%

“…Recently, in the VI image SR area, numerous methods [5][6][7][8][9][10][11][12][13][14][15][16][17][18] have been developed to find the most likely solutions. The existing VI SR methods can be roughly grouped into three categories: interpolation-based [5], prior-based [6], and learningbased methods [9][10][11][12][13][14][15][16][17][18][19][20]. Among them, the interpolation-based methods, such as the nearest neighbour, bilinear and bicubic interpolations, are the simplest and fastest.…”

Section: Introductionmentioning

confidence: 99%

Single infrared image super‐resolution based on lightweight multi‐path feature fusion network

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

Single infrared (IR) image super‐resolution methods can help to reduce the cost and difficulty in manufacturing IR sensors for the imaging system. However, the deep learning‐based image SR methods need to build a complex network and thus consume a lot of computational power, which limits the application of SR technology on devices with low computing resources in practice. To solve this problem, the authors present a lightweight multi‐path feature fusion network (MFFN) for the single infrared (IR) image SR. A multi‐path feature fusion block (MFFB) is developed to extract and fuse multiple and discriminative features in a recursive feedback way. Specifically, the multiple features are refined via the linear feature extraction branch, shared‐source residual feature extraction branch, and channel attention branch in MFFB. Finally, the authors reconstruct the high‐resolution IR images from the low‐resolution counterpart based on the refined multiple features. The experimental results demonstrate that MFFN achieves high‐quality single infrared image SR and shows superiority over previous methods for several scale factors (e.g. ×2, ×3, and ×4). MFFN has potential applications in the mobile infrared imaging system.

show abstract

Single image super-resolution reconstruction based on multi-scale feature mapping adversarial network

Cited by 20 publications

References 6 publications

Merging Super Resolution and Attribute Learning for Low-Resolution Person Attribute Recognition

Merging Super Resolution and Attribute Learning for Low-Resolution Person Attribute Recognition

A relic sketch extraction framework based on detail-aware hierarchical deep network

Single infrared image super‐resolution based on lightweight multi‐path feature fusion network

Contact Info

Product

Resources

About