Detecting Moving Humans Using Color and Infrared Video

We approach the task of human silhouette extraction from color and infrared video using automatic image reg-istration. Image registration between color and thermal im-ages is a challengingproblem due to the dificulties asso-ciated with finding correspondence. Howevel; the moving people in a static scene provide cues to address this prob-lem. In this papel; we propose a hierarchical scheme to automatically find the correspondence between the prelim-inary human silhouettes extracted from color and infrared video for image registration. Next, we discuss some strate-gies for probabilistically combining cues from registered color and thermal images. It is shown that the proposed ap-proach achieves good results for image registration and hu-man silhouette extraction. Experimental results also show a comparison of sensor fusion strategies and demonstrate the improvement in performance for human silhouette extrac-tion. 1