Abstract | ||
---|---|---|
•A probabilistic generative framework, PGM, is designed for image super-resolution.•The PGM assembles the advantages of coding-based and regression-based methods.•The PGM is developed with a conditional prior showing competitive performance.•The model has low computational cost and is robust to noise.•Three existing popular SR methods are shown to be reinvented under our framework. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1016/j.sigpro.2018.10.004 | Signal Processing |
Keywords | Field | DocType |
Probabilistic generative model,Image super-resolution,Conditional prior,Recognition model | Mathematical optimization,Parameterized complexity,Regression,Inference,Coding (social sciences),Artificial intelligence,Probabilistic logic,Generative grammar,Superresolution,Machine learning,Mathematics | Journal |
Volume | ISSN | Citations |
156 | 0165-1684 | 0 |
PageRank | References | Authors |
0.34 | 37 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Zhengjue Wang | 1 | 10 | 4.54 |
Bo Chen | 2 | 304 | 34.22 |
Hao Zhang | 3 | 203 | 64.03 |
Hongwei Liu | 4 | 376 | 63.93 |