Recovering missing data in three dimensional gait measurement

The main goal of our research is to construct a gait model that utilizes three information: three dimensional position of human body parts, myoelectric wave, and floor reaction pattern. This model aims at generating one of the three information when the other two information are given. We, however, encountered a problem in which the motion capture to contain maker occlusion. To solve for the relationship between the three data, we need obtain the non-missing data. This paper mainly focuses on interpolating the missing three dimensional position when the human body parts are out of the range of the motion capture camera. The result of the interpolation method to recover the best correct data is introduced and compared.