DCPNet: A Densely Connected Pyramid Network for Monocular Depth Estimation
Pyramid architecture is a useful strategy to fuse multi-scale features in deep monocular depth estimation approaches. However, most pyramid networks fuse features only within the adjacent stages in a pyramid structure. To take full advantage of the pyramid structure, inspired by the success of Dense...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-10-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/20/6780 |
| _version_ | 1797513234071158784 |
|---|---|
| author | Zhitong Lai Rui Tian Zhiguo Wu Nannan Ding Linjian Sun Yanjie Wang |
| author_facet | Zhitong Lai Rui Tian Zhiguo Wu Nannan Ding Linjian Sun Yanjie Wang |
| author_sort | Zhitong Lai |
| collection | DOAJ |
| description | Pyramid architecture is a useful strategy to fuse multi-scale features in deep monocular depth estimation approaches. However, most pyramid networks fuse features only within the adjacent stages in a pyramid structure. To take full advantage of the pyramid structure, inspired by the success of DenseNet, this paper presents DCPNet, a densely connected pyramid network that fuses multi-scale features from multiple stages of the pyramid structure. DCPNet not only performs feature fusion between the adjacent stages, but also non-adjacent stages. To fuse these features, we design a simple and effective dense connection module (DCM). In addition, we offer a new consideration of the common upscale operation in our approach. We believe DCPNet offers a more efficient way to fuse features from multiple scales in a pyramid-like network. We perform extensive experiments using both outdoor and indoor benchmark datasets (i.e., the KITTI and the NYU Depth V2 datasets) and DCPNet achieves the state-of-the-art results. |
| first_indexed | 2024-03-10T06:13:48Z |
| format | Article |
| id | doaj.art-dac7796296da4fdd9bf2829bec4e1ad3 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T06:13:48Z |
| publishDate | 2021-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-dac7796296da4fdd9bf2829bec4e1ad32023-11-22T19:57:21ZengMDPI AGSensors1424-82202021-10-012120678010.3390/s21206780DCPNet: A Densely Connected Pyramid Network for Monocular Depth EstimationZhitong Lai0Rui Tian1Zhiguo Wu2Nannan Ding3Linjian Sun4Yanjie Wang5Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaNational Space Science Center, Chinese Academy of Sciences, Beijing 100190, ChinaChangchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, Changchun 130033, ChinaPyramid architecture is a useful strategy to fuse multi-scale features in deep monocular depth estimation approaches. However, most pyramid networks fuse features only within the adjacent stages in a pyramid structure. To take full advantage of the pyramid structure, inspired by the success of DenseNet, this paper presents DCPNet, a densely connected pyramid network that fuses multi-scale features from multiple stages of the pyramid structure. DCPNet not only performs feature fusion between the adjacent stages, but also non-adjacent stages. To fuse these features, we design a simple and effective dense connection module (DCM). In addition, we offer a new consideration of the common upscale operation in our approach. We believe DCPNet offers a more efficient way to fuse features from multiple scales in a pyramid-like network. We perform extensive experiments using both outdoor and indoor benchmark datasets (i.e., the KITTI and the NYU Depth V2 datasets) and DCPNet achieves the state-of-the-art results.https://www.mdpi.com/1424-8220/21/20/6780monocular depth estimationpyramid networksdense connectionfeature fusion |
| spellingShingle | Zhitong Lai Rui Tian Zhiguo Wu Nannan Ding Linjian Sun Yanjie Wang DCPNet: A Densely Connected Pyramid Network for Monocular Depth Estimation Sensors monocular depth estimation pyramid networks dense connection feature fusion |
| title | DCPNet: A Densely Connected Pyramid Network for Monocular Depth Estimation |
| title_full | DCPNet: A Densely Connected Pyramid Network for Monocular Depth Estimation |
| title_fullStr | DCPNet: A Densely Connected Pyramid Network for Monocular Depth Estimation |
| title_full_unstemmed | DCPNet: A Densely Connected Pyramid Network for Monocular Depth Estimation |
| title_short | DCPNet: A Densely Connected Pyramid Network for Monocular Depth Estimation |
| title_sort | dcpnet a densely connected pyramid network for monocular depth estimation |
| topic | monocular depth estimation pyramid networks dense connection feature fusion |
| url | https://www.mdpi.com/1424-8220/21/20/6780 |
| work_keys_str_mv | AT zhitonglai dcpnetadenselyconnectedpyramidnetworkformonoculardepthestimation AT ruitian dcpnetadenselyconnectedpyramidnetworkformonoculardepthestimation AT zhiguowu dcpnetadenselyconnectedpyramidnetworkformonoculardepthestimation AT nannanding dcpnetadenselyconnectedpyramidnetworkformonoculardepthestimation AT linjiansun dcpnetadenselyconnectedpyramidnetworkformonoculardepthestimation AT yanjiewang dcpnetadenselyconnectedpyramidnetworkformonoculardepthestimation |