Understanding Camera Tracking

Before you start the process of tracking, it’s important that you understand how it works.

1. First, you load a video and track features using the tracking tools in Blender. Good features are typically details in the video that are highly visible, static relative to the world, and have a high contrast. Tracking uses these features found in the video to establish which way the objects in the video are moving relative to the frame. Although this is done frame by frame, it’s usually pretty automatic as Blender can track the features for you.

2. Next, when you have enough tracked markers in the video, you input the camera settings to tell Blender the type of camera and lens you used to film the video. If you don’t know exactly what lens you used, Blender will be able to estimate the settings and give you something pretty close to those you used.

3. Then, you have to address the camera’s movement. At this stage, Blender analyzes the points you tracked through the video and, by comparing their movements in different frames, it can reconstruct the camera’s perspective and determine where the 3D camera should be to see the points in those specific positions. That way, you’ll end up with a 3D camera that moves in exactly the same way as the real camera you used.

4. Finally, it’s just a matter of aligning the 3D camera and adjusting it to your scene so its orientation is correct and your 3D objects fit within the real footage.