This is the fun part. The training result, the model, will be used to swap the faces in the target video. Faceswap generates the output video in the form of a folder with the image sequences. You will need a tool to convert this to a video. The built-in tool to convert images to video didnt work for me. I used stop motion functionality from Corel VideoStudio. If the end results disappoints, its time to retrace steps in extraction or training. Converting is not as CPU/GPU intensive as training. You can at any point stop the training and try conversion out. Then when you start training again it builds on the last saved state of the model. If the model is crap, delete it and start over.
Since we do not know what effects Corel Gallery is generating, it is difficult to get a feel for the quality of the result. There are a lot of ways to check quality. I tried testing using images from the test or validation set. However, faces are much more dynamic then a single image. As a result, it was difficult to tell what condition the result was in as a whole. Faceswap is a tool in which you can view the result in real time. However, it was hard to tell if the performance is slow or not. If you generate a batch of images with another tool (for example, Corel VideoStudio), you can then convert those images to a video using faceswap. This will give you a good idea of how the output video will perform, should you ever want to save the output video to watch it.
Saving even more time, we can double the number of selections you can make using theMagic Wandby going into theSettingsdialog box and un-checking theUse the initial 20 imagesbox. This will allow you to click and drag over any part of the image to get the area you want to select. You can change the number of selection areas you can make just by clicking in the box in the bottom left of the dialogue and changing the number there. 3d9ccd7d82