Conference proceeding
Fast Single Image Reflection Suppression via Convex Optimization
2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vol.2019-, pp.8133-8141
06/2019
DOI: 10.1109/CVPR.2019.00833
Abstract
Removing undesired reflections from images taken through the glass is of great importance in computer vision. It serves as a means to enhance the image quality for aesthetic purposes as well as to preprocess images in machine learning and pattern recognition applications. We propose a convex model to suppress the reflection from a single input image. Our model implies a partial differential equation with gradient thresholding, which is solved efficiently using Discrete Cosine Transform. Extensive experiments on synthetic and real-world images demonstrate that our approach achieves desirable reflection suppression results and dramatically reduces the execution time.
Details
- Title: Subtitle
- Fast Single Image Reflection Suppression via Convex Optimization
- Creators
- Yang Yang - TencentWenye Ma - TencentYin Zheng - TencentJian-Feng Cai - Hong Kong University of Science and TechnologyWeiyu Xu - University of Iowa
- Resource Type
- Conference proceeding
- Publication Details
- 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Vol.2019-, pp.8133-8141
- Publisher
- IEEE
- DOI
- 10.1109/CVPR.2019.00833
- ISSN
- 1063-6919
- eISSN
- 2575-7075
- Language
- English
- Date published
- 06/2019
- Academic Unit
- Electrical and Computer Engineering
- Record Identifier
- 9984197432902771
Metrics
18 Record Views