GooderVideo. How SteadyHand Works. Step 2: Isolate Camera Motion

…make your good video even Gooder^™

How SteadyHand Works: Step 2:

Isolate Camera Motion from Scene Action

There is lots of movement among the billions of pixels in a video.

Some of the movement is due to action in the scene, but the vast majority of pixel movement is due to camera movement. This makes sense when you realize that one camera movement causes EVERY pixel to move.

SteadyHand takes in these millions and billions of pixel movements and applies artificial intelligence to sort out the camera movement from the movement in the scene.

In the images below, the tennis scene is analyzed to separate out the player movement from the camera pan movement.

In this image, only the action within the scene is used for the arrows.

From the arrows, we can see that the tennis player is crouching down and moving a little to the left.

The arrows in this image show the camera motion. In fact, it is only the horizontal component of the camera motion.

Notice how this image has VERY uniform arrows? This is different from the arrows in the previous web page (Step-1), where the arrows spread from the middle. Why? Because in Step-1, the arrows showed all the camera motion and the camera was zooming in on the player. This zoom made the image “grow larger.” That’s why the arrows (vectors) were all pointing away from the center.

The next step will tell more about how SteadyHand separates the different camera motions before correcting the scene.

I found the best welding camera and ultra dynamic range welding cameras for industrial imaging and high-dynamic range for arc welding camera