| Presentation | 2022-12-15 Single Image Raindrop Removal Using a Non-local Operator and Feature Maps in the Frequency Domain Shinya Ezumi, Masaaki Ikehara, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | High-quality raindrop removal is desired for outdoor image processing systems as well as for acquiring good-looking images. Existing methods often fail to keep global consistency and generate unnatural patterns. In this paper, we tackle this problem by introducing the non-local operator. The non-local operator combines features in distant locations with matrix multiplication and enables consistency in distant locations. In addition, we also propose to process feature maps in the frequency domain for more precise inpainting of high frequency components such as edges. Feature maps in the frequency domain are obtained by the FFT operation and processed by several convolution layers. Experimental results show that our method effectively removes raindrops and achieves state-of-the-art performance. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Deep Learning / Image Restoration / Raindrop Removal / Non-local Neural Network / Fast Fourier Transform |
| Paper # | PRMU2022-34 |
| Date of Issue | 2022-12-08 (PRMU) |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2022/12/15(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Toyama International Conference Center |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Seiichi Uchida(Kyushu Univ.) |
| Vice Chair | Takuya Funatomi(NAIST) / Mitsuru Anpai(Denso IT Lab.) |
| Secretary | Takuya Funatomi(CyberAgent) / Mitsuru Anpai(Univ. of Tokyo) |
| Assistant | Nakamasa Inoue(Tokyo Inst. of Tech.) / Yasutomo Kawanishi(Riken) |
| Paper Information | |
| Registration To | Technical Committee on Pattern Recognition and Media Understanding |
|---|---|
| Language | ENG-JTITLE |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Single Image Raindrop Removal Using a Non-local Operator and Feature Maps in the Frequency Domain |
| Sub Title (in English) | |
| Keyword(1) | Deep Learning |
| Keyword(2) | Image Restoration |
| Keyword(3) | Raindrop Removal |
| Keyword(4) | Non-local Neural Network |
| Keyword(5) | Fast Fourier Transform |
| 1st Author's Name | Shinya Ezumi |
| 1st Author's Affiliation | Keio University(Keio Univ.) |
| 2nd Author's Name | Masaaki Ikehara |
| 2nd Author's Affiliation | Keio University(Keio Univ.) |
| Date | 2022-12-15 |
| Paper # | PRMU2022-34 |
| Volume (vol) | vol.122 |
| Number (no) | PRMU-314 |
| Page | pp.pp.13-18(PRMU), |
| #Pages | 6 |
| Date of Issue | 2022-12-08 (PRMU) |