Q100179: How to fix banding issues when using the C_ColourMatcher

Follow

SYMPTOMS

When using the C_ColourMatcher node in CaraVR, you may notice that the result after analysis does not appear better than the initial solve, and you may also see a banding effect created.

As example, here is a problematic colour matching result:

 

CAUSE

C_ColourMatcher works by matching the whole of the camera input and if you have a round lens with a black border, then this can offset the colour matcher making the results worse.

 

RESOLUTION

To avoid the above issue you should:

  • Create an alpha for your camera inputs by rotoscoping around the image to avoid including any black areas, and point the C_CameraSolve node to use the alpha created above for the Mask shape parameter

  • Change the Mask Shape in the Cameras tab of the C_CameraSovler node to Elliptical, and reduce the size of the mask to remove the black areas

The C_ColourMatcher node will start using the resulting alpha when analysing the exposure and colours and will ignore the black part of the image during processing. This should help improve the C_ColourMatcher node's accuracy.

 

Additionally, CaraVR 1.0v2 introduced Multi-Band blending mode in the C_Stitcher node. Setting the Blend type to Multi-Band can also help reduce banding.

As example, here is the final colour matching result with fixes applied: 

NOTE:  All images are courtesy of Universal Post 

 

More information about the C_ColourMatcher node can be found in our documentation here: C_ColourMatcher

    We're sorry to hear that

    Please tell us why