Fast face detection and species classification of African great apes

Decline of biodiversity has become a serious problem on a global scale. Therefore environmental groups try to protect the endangered species populations. To this end, autonomous video traps are powerful instruments to monitor population sizes in the wild. These systems are used increasingly, however manually analyzing the huge amount of data gathered by video traps is taking a lot of time. Therefore automatic video analysis systems are in great demand. In this paper we show that technology developed for human face detection and analysis can successfully be applied to African great apes. We describe the algorithms and demonstrate the performance of the system using the example of chimpanzee and gorilla faces. Moreover we depict two different methods for species classification and compare the capabilities to distinguish between both species.