Provided are a method, system, and article of manufacture for generating a video output by combining a first video input and a second video input. A first input image is accessed from the first video input and a second input image is accessed from the second video input, wherein the second input image includes a selected element recorded in front of a monochromatic background. A first region and a second region are generated from the first input image. An output image that includes the first region and the second region with a gap separating the first region from the second region is generated. The selected element is placed at least over the gap in the output image, wherein predetermined areas of the first region and the second region remain visible in the output image. The output image is included in the video output.