SRResNet Performance Enhancement Using Patch Inputs and Partial Convolution-Based Padding
Source abstract
Abstract
Due to highly underdetermined nature of Single Image Super-Resolution (SISR) problem, deep learning neural networks are required to be more deeper to solve the problem effectively. One of deep neural networks successful in the Super-Resolution (SR) problem is ResNet which can render the capability of deeper networks with the help of skip connections. However, zero padding (ZP) scheme in the network restricts benefits of skip connections in SRResNet and its performance as the ratio of the number of pure input data to that of zero padded data increases. In this paper. we consider the ResNet with Partial Convolution based Padding (PCP) instead of ZP to solve SR problem. Since training of deep neural networks using patch images is advantageous in many aspects such as the number of training image data and network complexities, patch image based SR performance is compared with single full image based one. The experimental results show that patch based SRResNet SR results are better than single full image based ones and the performance of deep SRResNet with PCP is better than the one with ZP.
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Advanced Image Processing Techniques · Image and Signal Denoising Methods · Advanced Vision and Imaging
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (42)
Show source-provided references
1. GUN: Gradual upsampling network for single image super-resolution10.1109/ACCESS.2018.2855127 ↗
2. Learning from a handful volumes: MRI resolution enhancement with volumetric super-resolution forests
3. Depth estimation network for dual defocused images with different depth-of-field
4. Accurate image super-resolution using very deep convolutional networks
5. Deeply-recursive convolutional network for image super-resolution
6. Image super-resolution via deep recursive residual network
7. A review of machine learning and deep learning applications
8. Very deep convolutional networks for large-scale image recognition
9. Deep residual learning for image recognition
10. Identity mappings in deep residual networks
11. Photo-realistic single image super-resolution using a generative adversarial network
12. A deep journey into super-resolution: A survey10.1145/3390462 ↗
13. Learning a deep convolutional network for image supar-resolution
14. Enhanced deep residual networks for single image super-resolution
15. NTIRE 2017 challenge on single image super-resolution: Methods and results
16. Fast, accurate, and lightweight super-resolution with cascading residual network
17. Formresnet: Formatted residual learning for image restoration
18. Beyond a Gaussian denoiser: Residual learning of deep cnn for image denoising10.1109/TIP.2017.2662206 ↗
19. Image restoration using very deep convolutional encoder-decoder networks with symmetric skip connections
20. KOALAnet: Blind super-resolution using kernel-oriented adaptive local adjustment
21. Artificial intelligence-based semantic segmentation of ocular regions for biometrics and healthcare applications10.32604/cmc.2020.013249 ↗
22. Deep learning in DXA image segmentation10.32604/cmc.2021.013031 ↗
23. Fine-grained vehicle type classification using lightweight convolutional neural network with feature optimization and joint learning strategy10.1007/s11042-020-09171-3 ↗
24. A lightweight CNN based on transfer learning for COVID-19 diagnosis10.32604/cmc.2022.024589 ↗
25. AEI-DNET: A novel DenseNet model with an autoencoder for the stock market predictions using stock technical indicators10.3390/electronics11040611 ↗
26. Accurate segmentation of nuclear regions with multi-organ histopathology images using artificial intelligence for cancer diagnosis in personalized medicine10.3390/jpm11060515 ↗
27. Learning on the edge: Explicit boundary handling in CNNs
28. Cube padding for weakly-supervised saliency prediction in 360° videos
29. G. Liu, K. J. Shih, T. -C. Wang, F. A. Reda, K. Sapra et al., “Partial convolution based padding,” Technical report, NVIDIA Corporation, preprint arXiv:1811.11718, 2018. [Online] Available: http://arxiv.org/abs/1811.11718.
30. Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification
31. Batch normalization: Accelerating deep network training by reducing internal covariate shift
32. ImageNet large scale visual recognition challenge
33. Low-complexity single-image super-resolution based on nonnegative neighbor embedding
34. Zeyde · Curves and Surfaces, Lecture Notes in Computer Science · 2010 · 6920 · 71110.1007/978-3-642-27413-8_47 ↗
35. A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics
36. Super-resolution of sentinel-2 images: Learning a globally applicable deep neural network10.1016/j.isprsjprs.2018.09.018 ↗
37. Image quality assessment: From error measurement to structural similarity10.1109/TIP.2003.819861 ↗
38. FSIM: A feature similarity index for image quality assessment’10.1109/TIP.2011.2109730 ↗
39. Design of a hybrid measure for image similarity: A statistical, algebraic and information-theoretic approach10.1080/22797254.2019.1628617 ↗
40. A universal image quality index10.1109/97.995823 ↗
41. Image quality assessment through FSIM, SSIM, MSE and PSNR–a comparative study10.4236/jcc.2019.73002 ↗
42. Survey paper: Image quality assessment