We are doing an comparison among existing face recognition algorithms and approaches with related to 1 to N searching face recognition on very large data sets. Our aim is to improve the performance of face recognition algorithms to improve the performance to be used in very large data-sets of images and give accurate results.
We were unable to find literature on this specific topic which offers a quantitative comparison of these. As stated in related literature here are 2 approaches in face recognition,
1. Feature base
2. Holistic approach
We are trying to stay in the domain of Holistic approach and we are expecting to do a complete Algorithm analysis on popular and significantly accepted as better algorithms and do improvements to algorithms and try to improve the performance.