초록

용어보기논문에서 용어와 풀이말을 자동 추출한 결과로, 시범 서비스 중입니다. AI-Helper

Abstract ▼ AI-Helper

Estimating person's head pose is an essential problem for many applications and face analysis. It is hard to design a consistent feature descriptor that works for every person because there could be lots of differences for each individuals. Our approach to solve this problem is via 'Random Forests' ...

Estimating person's head pose is an essential problem for many applications and face analysis. It is hard to design a consistent feature descriptor that works for every person because there could be lots of differences for each individuals. Our approach to solve this problem is via 'Random Forests' which designs a random feature and have been shown an impressive result for classification and regression. One have to specify random questions of the nodes of decision trees which builds random forests to use the algorithm. In this paper, several features and several structure of random forests are considered to achieve the best performance. To do this, we find the way to deal with random forests which is consist of asymmetric decision trees. We achieved 86.5% accuracy with QMUL head pose dataset which has low resolution of two dimensional head images.

주제어