Abstract | ||
---|---|---|
This paper presents a novel multi-level generative chaotic Recurrent Neural Network (RNN) for image inpainting. This technique utilizes a general framework with multiple chaotic RNN that makes learning the image prior from a single corrupted image more robust and efficient. The proposed network utilizes a randomly-initialized process for parameterization, along with a unique quad-directional encoder structure, chaotic state transition, and adaptive importance for multi-level RNN updating. The efficacy of the approach has been validated through multiple experiments. In spite of a much lower computational load, quantitative comparisons reveal that the proposed approach exceeds the performance of several image-restoration benchmarks. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/WACV48630.2021.00367 | 2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION WACV 2021 |
DocType | ISSN | Citations |
Conference | 2472-6737 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Cong Chen | 1 | 38 | 12.03 |
Amos Lynn Abbott | 2 | 0 | 0.34 |
Daniel J. Stilwell | 3 | 0 | 0.34 |