Loading…
Assembling convolution neural networks for automatic viewing transformation
Images taken under different camera poses are rotated or distorted, which leads to poor perception experiences. This paper proposes a new framework to automatically transform the images to the conformable view setting by assembling different convolution neural networks. Specifically, a referential 3...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Default Article |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/2134/9912980.v1 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1818168067204579328 |
|---|---|
| author | Haibin Cai Lei Jiang Bangli Liu Yiqi Deng Qinggang Meng |
| author_facet | Haibin Cai Lei Jiang Bangli Liu Yiqi Deng Qinggang Meng |
| author_sort | Haibin Cai (794409) |
| collection | Figshare |
| description | Images taken under different camera poses are rotated or distorted, which leads to poor perception experiences. This paper proposes a new framework to automatically transform the images to the conformable view setting by assembling different convolution neural networks. Specifically, a referential 3D ground plane is firstly derived from the RGB image and a novel projection mapping algorithm is developed to achieve automatic viewing transformation. Extensive experimental results demonstrate that the proposed method outperforms the state-ofthe-art vanishing points based methods by a large margin in terms of accuracy and robustness. |
| format | Default Article |
| id | rr-article-9912980 |
| institution | Loughborough University |
| publishDate | 2019 |
| record_format | Figshare |
| spelling | rr-article-99129802019-09-09T00:00:00Z Assembling convolution neural networks for automatic viewing transformation Haibin Cai (794409) Lei Jiang (73366) Bangli Liu (6178787) Yiqi Deng (6004352) Qinggang Meng (1257072) Electrical & Electronic Engineering Information and Computing Sciences Engineering Technology Automatic viewing transform Convolution neural networks Deep learning Images taken under different camera poses are rotated or distorted, which leads to poor perception experiences. This paper proposes a new framework to automatically transform the images to the conformable view setting by assembling different convolution neural networks. Specifically, a referential 3D ground plane is firstly derived from the RGB image and a novel projection mapping algorithm is developed to achieve automatic viewing transformation. Extensive experimental results demonstrate that the proposed method outperforms the state-ofthe-art vanishing points based methods by a large margin in terms of accuracy and robustness. 2019-09-09T00:00:00Z Text Journal contribution 2134/9912980.v1 https://figshare.com/articles/journal_contribution/Assembling_convolution_neural_networks_for_automatic_viewing_transformation/9912980 All Rights Reserved |
| spellingShingle | Electrical & Electronic Engineering Information and Computing Sciences Engineering Technology Automatic viewing transform Convolution neural networks Deep learning Haibin Cai Lei Jiang Bangli Liu Yiqi Deng Qinggang Meng Assembling convolution neural networks for automatic viewing transformation |
| title | Assembling convolution neural networks for automatic viewing transformation |
| title_full | Assembling convolution neural networks for automatic viewing transformation |
| title_fullStr | Assembling convolution neural networks for automatic viewing transformation |
| title_full_unstemmed | Assembling convolution neural networks for automatic viewing transformation |
| title_short | Assembling convolution neural networks for automatic viewing transformation |
| title_sort | assembling convolution neural networks for automatic viewing transformation |
| topic | Electrical & Electronic Engineering Information and Computing Sciences Engineering Technology Automatic viewing transform Convolution neural networks Deep learning |
| url | https://hdl.handle.net/2134/9912980.v1 |