On 18/07/2016 at 09:13, xxxxxxxx wrote:
I have been searching for info about the Motion Vector pass, trying to understand what type of information is stored in the RGB values but it seems to be some sort of a secret.
Can anyone point me in the right direction?
Any information related with Motion Vectors is welcomed.
On 19/07/2016 at 01:44, xxxxxxxx wrote:
Hello,
actually you find information on the Motion Vector Multi-Pass in the Cinema 4D documentation.
Best wishes,
Sebastian
On 19/07/2016 at 06:35, xxxxxxxx wrote:
Well, that helped a bit.
I assume that values between 0 and 128 map from -1 to 0 and values from 128 to 255 map from 0 to 1, for the red and green channels, right?
Then, the fractional value will be used to generate a vector that defines the direction that the pixel must be dragged while loosing opacity.
And a multiplier can be applied to vary the amount to "dragging/blurring".
Am I assuming correctly?
On 25/07/2016 at 09:04, xxxxxxxx wrote:
Hello,
I got some information on this topic form the developers. The say that the motion vector pass follows the so-called RSMB encoding which is described on these pages:
http://revisionfx.com/faq/motion_vector/
http://revisionfx.com/faq/motion_vector_math/
best wishes,
Sebastian