Face Recognition by Elastic Bunch Graph Matching
IEEE Transactions on Pattern Analysis and Machine Intelligence
Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns
IEEE Transactions on Pattern Analysis and Machine Intelligence
Exploiting "approximate communication" for mobile media applications
Proceedings of the 10th workshop on Mobile Computing Systems and Applications
Mobile image recognition: architectures and tradeoffs
Proceedings of the Eleventh Workshop on Mobile Computing Systems & Applications
| Hi-index | 0.00 |
Face recognition has been extensively explored in diversified applications on ubiquitous devices. Most of the research has been primarily focused on full frontal/profile of facial images while proposing novice techniques to pursue this problem. The resource constraint in Mobile devices adds more complexity to the face recognition process. To reduce computational requirements some investigations are made to use the partial faces for recognition process. However, the inadequate information in partial faces makes the problem much more challenging and therefore limited attempts have been made in this direction. Our Active pixel based approach is capable of recognizing the persons using either full or partial face information. The technique reduces the computational resources compared to the LBP which was claimed as one of the most suitable approach on mobile devices. We have carried out the experiments on the YALE facial databases. Other works [1, 2] have used 50% vertical portion and showed the accuracy of correct recognition 94% within best five matches. In our dynamic partial matching we have used 10% to 34% image and obtained correct recognition rate from 96% to 100% within best three matches.