Reflection removal using ghosting cues
Photographs taken through glass windows often contain both the desired scene and undesired reflections. Separating the reflection and transmission layers is an important but ill-posed problem that has both aesthetic and practical applications. In this work, we introduce the use of ghosting cues that...
| Main Authors: | , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | en_US |
| Published: |
Institute of Electrical and Electronics Engineers (IEEE)
2018
|
| Online Access: | http://hdl.handle.net/1721.1/113822 https://orcid.org/0000-0001-9919-069X https://orcid.org/0000-0002-2231-7995 |
| _version_ | 1811089887875563520 |
|---|---|
| author | Krishnan, Dilip Shih, YiChang Durand, Frederic Freeman, William T. |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Krishnan, Dilip Shih, YiChang Durand, Frederic Freeman, William T. |
| author_sort | Krishnan, Dilip |
| collection | MIT |
| description | Photographs taken through glass windows often contain both the desired scene and undesired reflections. Separating the reflection and transmission layers is an important but ill-posed problem that has both aesthetic and practical applications. In this work, we introduce the use of ghosting cues that exploit asymmetry between the layers, thereby helping to reduce the ill-posedness of the problem. These cues arise from shifted double reflections of the reflected scene off the glass surface. In double-pane windows, each pane reflects shifted and attenuated versions of objects on the same side of the glass as the camera. For single-pane windows, ghosting cues arise from shifted reflections on the two surfaces of the glass pane. Even though the ghosting is sometimes barely perceptible by humans, we can still exploit the cue for layer separation. In this work, we model the ghosted reflection using a double-impulse convolution kernel, and automatically estimate the spatial separation and relative attenuation of the ghosted reflection components. To separate the layers, we propose an algorithm that uses a Gaussian Mixture Model for regularization. Our method is automatic and requires only a single input image. We demonstrate that our approach removes a large fraction of reflections on both synthetic and real-world inputs. |
| first_indexed | 2024-09-23T14:26:35Z |
| format | Article |
| id | mit-1721.1/113822 |
| institution | Massachusetts Institute of Technology |
| language | en_US |
| last_indexed | 2024-09-23T14:26:35Z |
| publishDate | 2018 |
| publisher | Institute of Electrical and Electronics Engineers (IEEE) |
| record_format | dspace |
| spelling | mit-1721.1/1138222022-09-29T09:29:13Z Reflection removal using ghosting cues Krishnan, Dilip Shih, YiChang Durand, Frederic Freeman, William T. Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science Shih, YiChang Durand, Frederic Freeman, William T. Photographs taken through glass windows often contain both the desired scene and undesired reflections. Separating the reflection and transmission layers is an important but ill-posed problem that has both aesthetic and practical applications. In this work, we introduce the use of ghosting cues that exploit asymmetry between the layers, thereby helping to reduce the ill-posedness of the problem. These cues arise from shifted double reflections of the reflected scene off the glass surface. In double-pane windows, each pane reflects shifted and attenuated versions of objects on the same side of the glass as the camera. For single-pane windows, ghosting cues arise from shifted reflections on the two surfaces of the glass pane. Even though the ghosting is sometimes barely perceptible by humans, we can still exploit the cue for layer separation. In this work, we model the ghosted reflection using a double-impulse convolution kernel, and automatically estimate the spatial separation and relative attenuation of the ghosted reflection components. To separate the layers, we propose an algorithm that uses a Gaussian Mixture Model for regularization. Our method is automatic and requires only a single input image. We demonstrate that our approach removes a large fraction of reflections on both synthetic and real-world inputs. Quanta Computer (Firm) Qatar Computing Research Institute 2018-02-20T14:42:52Z 2018-02-20T14:42:52Z 2015-10 2015-06 Article http://purl.org/eprint/type/ConferencePaper 978-1-4673-6964-0 http://hdl.handle.net/1721.1/113822 YiChang Shih, et al. "Reflection Removal Using Ghosting Cues." 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 7-12 June 2015, Boston, Massachusetts, IEEE, 2015, pp. 3193–201. https://orcid.org/0000-0001-9919-069X https://orcid.org/0000-0002-2231-7995 en_US http://dx.doi.org/10.1109/CVPR.2015.7298939 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Institute of Electrical and Electronics Engineers (IEEE) MIT Web Domain |
| spellingShingle | Krishnan, Dilip Shih, YiChang Durand, Frederic Freeman, William T. Reflection removal using ghosting cues |
| title | Reflection removal using ghosting cues |
| title_full | Reflection removal using ghosting cues |
| title_fullStr | Reflection removal using ghosting cues |
| title_full_unstemmed | Reflection removal using ghosting cues |
| title_short | Reflection removal using ghosting cues |
| title_sort | reflection removal using ghosting cues |
| url | http://hdl.handle.net/1721.1/113822 https://orcid.org/0000-0001-9919-069X https://orcid.org/0000-0002-2231-7995 |
| work_keys_str_mv | AT krishnandilip reflectionremovalusingghostingcues AT shihyichang reflectionremovalusingghostingcues AT durandfrederic reflectionremovalusingghostingcues AT freemanwilliamt reflectionremovalusingghostingcues |