Birleşik olarak hareket yakalama verilerinin filtrelenmesi ve ana çerçevelerin azaltılması

Two new methods for combined filtering and key-frame reduction of motioncapture data are proposed. Filtering of motion capture data is necessary toeliminate any jitter introduced by a motion capture system. Although jitterremoval is needed to obtain a more realistic animation, it may result in an over-smoothed motion data if it is not done properly. Key-frame reduction, on theother hand, allows animators to easily edit motion data by representing animationcurves with a significantly smaller number of key frames. One of the proposedtechniques achieves key frame reduction and jitter removal simultaneously byfitting a Hermite curve to motion capture data using dynamic programming.Another method is to use curve simplification algorithms on the motion capturedata until the desired reduction is reached. In this research, the results of thesealgorithms are evaluated and compared. Both subjective and objective resultsare presented.Keywords: Motion Capture, Keyframe Reduction, Curve Fitting, Curve Simpli-fication, Noise Filtering.iv