This is a legacy university project (2011) meant to test the potential and performances of an android application implementing face recognition through the openCV library. It's now outdated due to the availability of more performing new libraries, and also the results are to be considered an historical view, considering the computational power at disposal of modern smartphones.
Camera Assistant application - Lorenzo Cavazzi - 754987
POLITECNICO DI MILANO - Image analysis course - AA 2011
Supervisor: Alessandro Giusti
Contacts: [email protected] / [email protected]
This archive contains all the source code, the binaries and a detailed report of project, plus 2 demostrative videos. Target platform is android 2.3+ and source code requires "openCV 2.3.1 for android" library in order to compile.
The content is divived in folders as follows: /doc: report in pdf with project details. /bin: binaries runnable from an android phone. To install application see instruction in /bin/README. /video: contains 2 videos, first one is a demostration of the application capabilities and its results in optimal condition, while the latter focuses on some problems discussed in the report. /src: source code. See instruction in /src/README to properly set development environment and how to compile
Experimental files aren't included, but demonstration videos are provided instead. Should you be willing to test the application firsthand, please install it and try to target peoples or people's photos.