Light‐field image super‐resolution with depth feature by multiple‐decouple and fusion module
Abstract Light‐field (LF) images offer the potential to improve feature capture in live scenes from multiple perspectives, and also generate additional normal vectors for performing super‐resolution (SR) image processing. With the benefit of machine learning, established AI‐based deep CNN models for...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-01-01
|
| Series: | Electronics Letters |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ell2.13019 |
| _version_ | 1827385184660488192 |
|---|---|
| author | Ka‐Hou Chan Sio‐Kei Im |
| author_facet | Ka‐Hou Chan Sio‐Kei Im |
| author_sort | Ka‐Hou Chan |
| collection | DOAJ |
| description | Abstract Light‐field (LF) images offer the potential to improve feature capture in live scenes from multiple perspectives, and also generate additional normal vectors for performing super‐resolution (SR) image processing. With the benefit of machine learning, established AI‐based deep CNN models for LF image SR often individualize the models for various resolutions. However, the rigidity of these approaches for actual LF applications stems from the considerable diversity in angular resolution among LF instruments. Therefore, an advanced neural network proposal is required to utilize a CNN‐based model for super‐resolving LF images with different resolutions obtained from provided features. In this work, a preprocessing to calculate the depth channel from given LF information is first presented, and then a multiple‐decouple and fusion module is introduced to integrate the VGGreNet for the LF image SR, which collects global‐to‐local information according to the CNN kernel size and dynamically constructs each view through a global view module. Besides, the generated features are transformed to a uniform space to perform final fusion, achieving global alignment for precise extraction of angular information. Experimental results show that the proposed method can handle benchmark LF datasets with various angular and different resolutions, reporting the effectiveness and potential performance of the method. |
| first_indexed | 2024-03-08T15:16:29Z |
| format | Article |
| id | doaj.art-a530e80e900f4fcab6e65d8a96f355fc |
| institution | Directory Open Access Journal |
| issn | 0013-5194 1350-911X |
| language | English |
| last_indexed | 2024-03-08T15:16:29Z |
| publishDate | 2024-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Electronics Letters |
| spelling | doaj.art-a530e80e900f4fcab6e65d8a96f355fc2024-01-10T11:00:34ZengWileyElectronics Letters0013-51941350-911X2024-01-01601n/an/a10.1049/ell2.13019Light‐field image super‐resolution with depth feature by multiple‐decouple and fusion moduleKa‐Hou Chan0Sio‐Kei Im1Faculty of Applied Sciences Macao Polytechnic University Macau ChinaFaculty of Applied Sciences Macao Polytechnic University Macau ChinaAbstract Light‐field (LF) images offer the potential to improve feature capture in live scenes from multiple perspectives, and also generate additional normal vectors for performing super‐resolution (SR) image processing. With the benefit of machine learning, established AI‐based deep CNN models for LF image SR often individualize the models for various resolutions. However, the rigidity of these approaches for actual LF applications stems from the considerable diversity in angular resolution among LF instruments. Therefore, an advanced neural network proposal is required to utilize a CNN‐based model for super‐resolving LF images with different resolutions obtained from provided features. In this work, a preprocessing to calculate the depth channel from given LF information is first presented, and then a multiple‐decouple and fusion module is introduced to integrate the VGGreNet for the LF image SR, which collects global‐to‐local information according to the CNN kernel size and dynamically constructs each view through a global view module. Besides, the generated features are transformed to a uniform space to perform final fusion, achieving global alignment for precise extraction of angular information. Experimental results show that the proposed method can handle benchmark LF datasets with various angular and different resolutions, reporting the effectiveness and potential performance of the method.https://doi.org/10.1049/ell2.13019adaptive signal processingimage fusionimage processingneural net architecturespatial filters |
| spellingShingle | Ka‐Hou Chan Sio‐Kei Im Light‐field image super‐resolution with depth feature by multiple‐decouple and fusion module Electronics Letters adaptive signal processing image fusion image processing neural net architecture spatial filters |
| title | Light‐field image super‐resolution with depth feature by multiple‐decouple and fusion module |
| title_full | Light‐field image super‐resolution with depth feature by multiple‐decouple and fusion module |
| title_fullStr | Light‐field image super‐resolution with depth feature by multiple‐decouple and fusion module |
| title_full_unstemmed | Light‐field image super‐resolution with depth feature by multiple‐decouple and fusion module |
| title_short | Light‐field image super‐resolution with depth feature by multiple‐decouple and fusion module |
| title_sort | light field image super resolution with depth feature by multiple decouple and fusion module |
| topic | adaptive signal processing image fusion image processing neural net architecture spatial filters |
| url | https://doi.org/10.1049/ell2.13019 |
| work_keys_str_mv | AT kahouchan lightfieldimagesuperresolutionwithdepthfeaturebymultipledecoupleandfusionmodule AT siokeiim lightfieldimagesuperresolutionwithdepthfeaturebymultipledecoupleandfusionmodule |