Abstract | ||
---|---|---|
In patch based face super-resolution method, the patch size is usually very small, and neighbor patches' relationship via overlapped regions is only to keep smoothness of reconstructed high-resolution image, so the prior is not always strong enough to regularize super-resolution when observed low-resolution image lose facial structure information. We propose to use Gaussian Mixture Model(GMM) to learn facial prior embedded between un-overlapped regions of neighbor patches. This approach, which has never been used to regularize face super-resolution before, usually works as a potential function in 8-connected Markov Random Fields (MRFs) with belief propagation. In the proposed algorithm, we assign high probability to the neighbor candidate patches that express correct facial structure, and others not. Experiments demonstrate that our method is superior in preserving smoothness and recovers facial structure and local details when low-resolution image lost the details of facial structure. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1109/ICPR.2008.4761426 | 19TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOLS 1-6 |
Keywords | Field | DocType |
pixel,gaussian mixture model,markov processes,super resolution,face,random processes,databases,low resolution,gaussian processes,face recognition,belief propagation,image resolution,mathematical model | Computer vision,Facial recognition system,Markov process,Random field,Pattern recognition,Computer science,Markov chain,Stochastic process,Artificial intelligence,Gaussian process,Mixture model,Belief propagation | Conference |
ISSN | Citations | PageRank |
1051-4651 | 4 | 0.41 |
References | Authors | |
5 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Kai Guo | 1 | 15 | 1.94 |
Xiaokang Yang | 2 | 3581 | 238.09 |
Rui Zhang | 3 | 92 | 7.65 |
Guangtao Zhai | 4 | 1707 | 145.33 |
Yu Song | 5 | 356 | 52.74 |