Multi-scale image harmonization
© 2010 ACM. Traditional image compositing techniques, such as alpha matting and gradient domain compositing, are used to create composites that have plausible boundaries. But when applied to images taken from different sources or shot under different conditions, these techniques can produce unrealis...
| Những tác giả chính: | , , , |
|---|---|
| Tác giả khác: | |
| Định dạng: | Bài viết |
| Ngôn ngữ: | English |
| Được phát hành: |
Association for Computing Machinery (ACM)
2021
|
| Truy cập trực tuyến: | https://hdl.handle.net/1721.1/134252 |
| _version_ | 1826194508087296000 |
|---|---|
| author | Sunkavalli, Kalyan Johnson, Micah K Matusik, Wojciech Pfister, Hanspeter |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Sunkavalli, Kalyan Johnson, Micah K Matusik, Wojciech Pfister, Hanspeter |
| author_sort | Sunkavalli, Kalyan |
| collection | MIT |
| description | © 2010 ACM. Traditional image compositing techniques, such as alpha matting and gradient domain compositing, are used to create composites that have plausible boundaries. But when applied to images taken from different sources or shot under different conditions, these techniques can produce unrealistic results. In this work, we present a framework that explicitly matches the visual appearance of images through a process we call image harmonization, before blending them. At the heart of this framework is a multi-scale technique that allows us to transfer the appearance of one image to another. We show that by carefully manipulating the scales of a pyramid decomposition of an image, we can match contrast, texture, noise, and blur, while avoiding image artifacts. The output composite can then be reconstructed from the modified pyramid coefficients while enforcing both alpha-based and seamless boundary constraints. We show how the proposed framework can be used to produce realistic composites with minimal user interaction in a number of different scenarios. |
| first_indexed | 2024-09-23T09:57:11Z |
| format | Article |
| id | mit-1721.1/134252 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T09:57:11Z |
| publishDate | 2021 |
| publisher | Association for Computing Machinery (ACM) |
| record_format | dspace |
| spelling | mit-1721.1/1342522023-02-28T20:30:29Z Multi-scale image harmonization Sunkavalli, Kalyan Johnson, Micah K Matusik, Wojciech Pfister, Hanspeter Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory © 2010 ACM. Traditional image compositing techniques, such as alpha matting and gradient domain compositing, are used to create composites that have plausible boundaries. But when applied to images taken from different sources or shot under different conditions, these techniques can produce unrealistic results. In this work, we present a framework that explicitly matches the visual appearance of images through a process we call image harmonization, before blending them. At the heart of this framework is a multi-scale technique that allows us to transfer the appearance of one image to another. We show that by carefully manipulating the scales of a pyramid decomposition of an image, we can match contrast, texture, noise, and blur, while avoiding image artifacts. The output composite can then be reconstructed from the modified pyramid coefficients while enforcing both alpha-based and seamless boundary constraints. We show how the proposed framework can be used to produce realistic composites with minimal user interaction in a number of different scenarios. 2021-10-27T20:04:11Z 2021-10-27T20:04:11Z 2010 2019-06-21T13:46:51Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/134252 Sunkavalli, K., et al. "Multi-Scale Image Harmonization." Acm Transactions on Graphics 29 4 (2010). en 10.1145/1778765.1778862 ACM Transactions on Graphics Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf Association for Computing Machinery (ACM) MIT web domain |
| spellingShingle | Sunkavalli, Kalyan Johnson, Micah K Matusik, Wojciech Pfister, Hanspeter Multi-scale image harmonization |
| title | Multi-scale image harmonization |
| title_full | Multi-scale image harmonization |
| title_fullStr | Multi-scale image harmonization |
| title_full_unstemmed | Multi-scale image harmonization |
| title_short | Multi-scale image harmonization |
| title_sort | multi scale image harmonization |
| url | https://hdl.handle.net/1721.1/134252 |
| work_keys_str_mv | AT sunkavallikalyan multiscaleimageharmonization AT johnsonmicahk multiscaleimageharmonization AT matusikwojciech multiscaleimageharmonization AT pfisterhanspeter multiscaleimageharmonization |