Abstract: A process for integrating video images includes generating a dancer video with a dancer and identifying a dancer head position of the dancer in the dancer video. A participant head location is identified and a cropped area is generated based on the participant head location. A participant video is generated of cropped area contents and a composite video is generated by overlaying the cropped area contents onto the dancer head position. A video image integration system includes a video camera for capturing participant videos and a processor. The processor has a processor database module storing pre-recorded dancer videos, a face recognition module configured to identify participant head locations from the video camera and generated cropped areas based thereon, and a blender module configured to overlay the cropped areas onto the dancer videos to generate composite videos.