Real-Time Single Image Depth Perception in the Wild with Handheld Devices
Depth perception is paramount for tackling real-world problems, ranging from autonomous driving to consumer applications. For the latter, depth estimation from a single image would represent the most versatile solution since a standard camera is available on almost any handheld device. Nonetheless,...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/1/15 |
| _version_ | 1827699485771300864 |
|---|---|
| author | Filippo Aleotti Giulio Zaccaroni Luca Bartolomei Matteo Poggi Fabio Tosi Stefano Mattoccia |
| author_facet | Filippo Aleotti Giulio Zaccaroni Luca Bartolomei Matteo Poggi Fabio Tosi Stefano Mattoccia |
| author_sort | Filippo Aleotti |
| collection | DOAJ |
| description | Depth perception is paramount for tackling real-world problems, ranging from autonomous driving to consumer applications. For the latter, depth estimation from a single image would represent the most versatile solution since a standard camera is available on almost any handheld device. Nonetheless, two main issues limit the practical deployment of monocular depth estimation methods on such devices: (i) the low reliability when deployed in the wild and (ii) the resources needed to achieve real-time performance, often not compatible with low-power embedded systems. Therefore, in this paper, we deeply investigate all these issues, showing how they are both addressable by adopting appropriate network design and training strategies. Moreover, we also outline how to map the resulting networks on handheld devices to achieve real-time performance. Our thorough evaluation highlights the ability of such fast networks to generalize well to new environments, a crucial feature required to tackle the extremely varied contexts faced in real applications. Indeed, to further support this evidence, we report experimental results concerning real-time, depth-aware augmented reality and image blurring with smartphones in the wild. |
| first_indexed | 2024-03-10T13:52:22Z |
| format | Article |
| id | doaj.art-83e5aac131de4abd8bf18ee9b2d826a8 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T13:52:22Z |
| publishDate | 2020-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-83e5aac131de4abd8bf18ee9b2d826a82023-11-21T02:02:27ZengMDPI AGSensors1424-82202020-12-012111510.3390/s21010015Real-Time Single Image Depth Perception in the Wild with Handheld DevicesFilippo Aleotti0Giulio Zaccaroni1Luca Bartolomei2Matteo Poggi3Fabio Tosi4Stefano Mattoccia5Department of Computer Science and Engineering, University of Bologna, 40136 Bologna, ItalyDepartment of Computer Science and Engineering, University of Bologna, 40136 Bologna, ItalyDepartment of Computer Science and Engineering, University of Bologna, 40136 Bologna, ItalyDepartment of Computer Science and Engineering, University of Bologna, 40136 Bologna, ItalyDepartment of Computer Science and Engineering, University of Bologna, 40136 Bologna, ItalyDepartment of Computer Science and Engineering, University of Bologna, 40136 Bologna, ItalyDepth perception is paramount for tackling real-world problems, ranging from autonomous driving to consumer applications. For the latter, depth estimation from a single image would represent the most versatile solution since a standard camera is available on almost any handheld device. Nonetheless, two main issues limit the practical deployment of monocular depth estimation methods on such devices: (i) the low reliability when deployed in the wild and (ii) the resources needed to achieve real-time performance, often not compatible with low-power embedded systems. Therefore, in this paper, we deeply investigate all these issues, showing how they are both addressable by adopting appropriate network design and training strategies. Moreover, we also outline how to map the resulting networks on handheld devices to achieve real-time performance. Our thorough evaluation highlights the ability of such fast networks to generalize well to new environments, a crucial feature required to tackle the extremely varied contexts faced in real applications. Indeed, to further support this evidence, we report experimental results concerning real-time, depth-aware augmented reality and image blurring with smartphones in the wild.https://www.mdpi.com/1424-8220/21/1/15monocular depth estimationdeep learningmobile systemssmartphone |
| spellingShingle | Filippo Aleotti Giulio Zaccaroni Luca Bartolomei Matteo Poggi Fabio Tosi Stefano Mattoccia Real-Time Single Image Depth Perception in the Wild with Handheld Devices Sensors monocular depth estimation deep learning mobile systems smartphone |
| title | Real-Time Single Image Depth Perception in the Wild with Handheld Devices |
| title_full | Real-Time Single Image Depth Perception in the Wild with Handheld Devices |
| title_fullStr | Real-Time Single Image Depth Perception in the Wild with Handheld Devices |
| title_full_unstemmed | Real-Time Single Image Depth Perception in the Wild with Handheld Devices |
| title_short | Real-Time Single Image Depth Perception in the Wild with Handheld Devices |
| title_sort | real time single image depth perception in the wild with handheld devices |
| topic | monocular depth estimation deep learning mobile systems smartphone |
| url | https://www.mdpi.com/1424-8220/21/1/15 |
| work_keys_str_mv | AT filippoaleotti realtimesingleimagedepthperceptioninthewildwithhandhelddevices AT giuliozaccaroni realtimesingleimagedepthperceptioninthewildwithhandhelddevices AT lucabartolomei realtimesingleimagedepthperceptioninthewildwithhandhelddevices AT matteopoggi realtimesingleimagedepthperceptioninthewildwithhandhelddevices AT fabiotosi realtimesingleimagedepthperceptioninthewildwithhandhelddevices AT stefanomattoccia realtimesingleimagedepthperceptioninthewildwithhandhelddevices |