Monocular Depth Estimation from a Single Infrared Image
Thermal infrared imaging is attracting much attention due to its strength against illuminance variation. However, because of the spectral difference between thermal infrared images and RGB images, the existing research on self-supervised monocular depth estimation has performance limitations. Theref...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/11/1729 |
| _version_ | 1797493743025127424 |
|---|---|
| author | Daechan Han Yukyung Choi |
| author_facet | Daechan Han Yukyung Choi |
| author_sort | Daechan Han |
| collection | DOAJ |
| description | Thermal infrared imaging is attracting much attention due to its strength against illuminance variation. However, because of the spectral difference between thermal infrared images and RGB images, the existing research on self-supervised monocular depth estimation has performance limitations. Therefore, in this study, we propose a novel Self-Guided Framework using a Pseudolabel predicted from RGB images. Our proposed framework, which solves the problem of appearance matching loss in the existing framework, transfers the high accuracy of Pseudolabel to the thermal depth estimation network by comparing low- and high-level pixels. Furthermore, we propose Patch-NetVLAD Loss, which strengthens local detail and global context information in the depth map from thermal infrared imaging by comparing locally global patch-level descriptors. Finally, we introduce an Image Matching Loss to estimate a more accurate depth map in a thermal depth network by enhancing the performance of the Pseudolabel. We demonstrate that the proposed framework shows significant performance improvement even when applied to various depth networks in the KAIST Multispectral Dataset. |
| first_indexed | 2024-03-10T01:24:22Z |
| format | Article |
| id | doaj.art-e499065aba7142e8a59dd3c15afc3221 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T01:24:22Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-e499065aba7142e8a59dd3c15afc32212023-11-23T13:54:56ZengMDPI AGElectronics2079-92922022-05-011111172910.3390/electronics11111729Monocular Depth Estimation from a Single Infrared ImageDaechan Han0Yukyung Choi1Department of Intelligent Mechatronics Engineering, Sejong University, Seoul 05006, KoreaDepartment of Intelligent Mechatronics Engineering, Sejong University, Seoul 05006, KoreaThermal infrared imaging is attracting much attention due to its strength against illuminance variation. However, because of the spectral difference between thermal infrared images and RGB images, the existing research on self-supervised monocular depth estimation has performance limitations. Therefore, in this study, we propose a novel Self-Guided Framework using a Pseudolabel predicted from RGB images. Our proposed framework, which solves the problem of appearance matching loss in the existing framework, transfers the high accuracy of Pseudolabel to the thermal depth estimation network by comparing low- and high-level pixels. Furthermore, we propose Patch-NetVLAD Loss, which strengthens local detail and global context information in the depth map from thermal infrared imaging by comparing locally global patch-level descriptors. Finally, we introduce an Image Matching Loss to estimate a more accurate depth map in a thermal depth network by enhancing the performance of the Pseudolabel. We demonstrate that the proposed framework shows significant performance improvement even when applied to various depth networks in the KAIST Multispectral Dataset.https://www.mdpi.com/2079-9292/11/11/1729monocular depth estimationself-supervised learninginfrared imagenight visionpseudo-labellocal descriptor |
| spellingShingle | Daechan Han Yukyung Choi Monocular Depth Estimation from a Single Infrared Image Electronics monocular depth estimation self-supervised learning infrared image night vision pseudo-label local descriptor |
| title | Monocular Depth Estimation from a Single Infrared Image |
| title_full | Monocular Depth Estimation from a Single Infrared Image |
| title_fullStr | Monocular Depth Estimation from a Single Infrared Image |
| title_full_unstemmed | Monocular Depth Estimation from a Single Infrared Image |
| title_short | Monocular Depth Estimation from a Single Infrared Image |
| title_sort | monocular depth estimation from a single infrared image |
| topic | monocular depth estimation self-supervised learning infrared image night vision pseudo-label local descriptor |
| url | https://www.mdpi.com/2079-9292/11/11/1729 |
| work_keys_str_mv | AT daechanhan monoculardepthestimationfromasingleinfraredimage AT yukyungchoi monoculardepthestimationfromasingleinfraredimage |