Extraction of Human Limb Regions and Parameter Estimation based on Curl of Optical Flow

We propose a method for extracting a rotating limb region and estimating its motion parameters, which is based on the curl of optical flow. Regarding the optical flow as a vector field, the proposed method computes the curl of the optical flow to extract a rotating region. The method assumes that one angular velocity varies in a short time period. By using the EM algorithm, the center of rotation along with rotation angles for each frame is estimated, and simultaneously the rotating region is segmented. Since the proposed method doesn’t require a simplicity of background, and can discriminate translational movements produced by other moving objects in the background, it is applicable to a practical situation in which conventional methods don’t work well. Experimental results with real image sequences demonstrate that the limb extraction can be performed well from noisy optical flow which are obtained from real images.