Deep Non-Line-of-Sight Imaging Using Echolocation
Non-line-of-sight (NLOS) imaging is aimed at visualizing hidden scenes from an observer’s (e.g., camera) viewpoint. Typically, hidden scenes are reconstructed using diffused signals that emit light sources using optical equipment and are reflected multiple times. Optical systems are commonly adopted...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/21/8477 |
| _version_ | 1797466442678927360 |
|---|---|
| author | Seungwoo Jang Ui-Hyeon Shin Kwangsu Kim |
| author_facet | Seungwoo Jang Ui-Hyeon Shin Kwangsu Kim |
| author_sort | Seungwoo Jang |
| collection | DOAJ |
| description | Non-line-of-sight (NLOS) imaging is aimed at visualizing hidden scenes from an observer’s (e.g., camera) viewpoint. Typically, hidden scenes are reconstructed using diffused signals that emit light sources using optical equipment and are reflected multiple times. Optical systems are commonly adopted in NLOS imaging because lasers can transport energy and focus light over long distances without loss. In contrast, we propose NLOS imaging using acoustic equipment inspired by echolocation. Existing acoustic NLOS is a computational method motivated by seismic imaging that analyzes the geometry of underground structures. However, this physical method is susceptible to noise and requires a clear signal, resulting in long data acquisition times. Therefore, we reduced the scan time by modifying the echoes to be collected simultaneously rather than sequentially. Then, we propose end-to-end deep-learning models to overcome the challenges of echoes interfering with each other. We designed three distinctive architectures: an encoder that extracts features by dividing multi-channel echoes into groups and merging them hierarchically, a generator that constructs an image of the hidden object, and a discriminator that compares the generated image with the ground-truth image. The proposed model successfully reconstructed the outline of the hidden objects. |
| first_indexed | 2024-03-09T18:39:57Z |
| format | Article |
| id | doaj.art-37f90ad3ee454c20827015e3714f4664 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T18:39:57Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-37f90ad3ee454c20827015e3714f46642023-11-24T06:48:39ZengMDPI AGSensors1424-82202022-11-012221847710.3390/s22218477Deep Non-Line-of-Sight Imaging Using EcholocationSeungwoo Jang0Ui-Hyeon Shin1Kwangsu Kim2Department of Artificial Intelligence, Sungkyunkwan University, Suwon 16419, KoreaDepartment of Artificial Intelligence, Sungkyunkwan University, Suwon 16419, KoreaCollege of Computing and Informatics, Sungkyunkwan University, Suwon 16419, KoreaNon-line-of-sight (NLOS) imaging is aimed at visualizing hidden scenes from an observer’s (e.g., camera) viewpoint. Typically, hidden scenes are reconstructed using diffused signals that emit light sources using optical equipment and are reflected multiple times. Optical systems are commonly adopted in NLOS imaging because lasers can transport energy and focus light over long distances without loss. In contrast, we propose NLOS imaging using acoustic equipment inspired by echolocation. Existing acoustic NLOS is a computational method motivated by seismic imaging that analyzes the geometry of underground structures. However, this physical method is susceptible to noise and requires a clear signal, resulting in long data acquisition times. Therefore, we reduced the scan time by modifying the echoes to be collected simultaneously rather than sequentially. Then, we propose end-to-end deep-learning models to overcome the challenges of echoes interfering with each other. We designed three distinctive architectures: an encoder that extracts features by dividing multi-channel echoes into groups and merging them hierarchically, a generator that constructs an image of the hidden object, and a discriminator that compares the generated image with the ground-truth image. The proposed model successfully reconstructed the outline of the hidden objects.https://www.mdpi.com/1424-8220/22/21/8477non-line-of-sightacoustic sensingdepth estimationdeep learning |
| spellingShingle | Seungwoo Jang Ui-Hyeon Shin Kwangsu Kim Deep Non-Line-of-Sight Imaging Using Echolocation Sensors non-line-of-sight acoustic sensing depth estimation deep learning |
| title | Deep Non-Line-of-Sight Imaging Using Echolocation |
| title_full | Deep Non-Line-of-Sight Imaging Using Echolocation |
| title_fullStr | Deep Non-Line-of-Sight Imaging Using Echolocation |
| title_full_unstemmed | Deep Non-Line-of-Sight Imaging Using Echolocation |
| title_short | Deep Non-Line-of-Sight Imaging Using Echolocation |
| title_sort | deep non line of sight imaging using echolocation |
| topic | non-line-of-sight acoustic sensing depth estimation deep learning |
| url | https://www.mdpi.com/1424-8220/22/21/8477 |
| work_keys_str_mv | AT seungwoojang deepnonlineofsightimagingusingecholocation AT uihyeonshin deepnonlineofsightimagingusingecholocation AT kwangsukim deepnonlineofsightimagingusingecholocation |