抄録
This paper presents a method for automated human detection using fisheye image. We introduce a probabilistic model to describe the wide variation of human appearance in hemispherical image. In our method, human is modeled as probabilistic templates of body silhouette and head-shoulder contour. These template features are extracted from the human images taken at various distance and orientation with respect to the camera, and form the training data set for model creation. A probabilistic appearance model is built by using the combination of principal component analysis (PCA) and kernel ridge regression (KRR). Finally, the problem of human detection is formulated as maximum a posteriori (MAP) estimation using above model. Experiments are conducted on indoor space where a fisheye lens camera is installed on the ceiling of crossing hallway. The feasibility and accuracy of our method is discussed through the experimental results.
