Unfortunately you can't do so.
There are some functions in the Vision Lib that only accept 8bit images. In order to use them you have to convert your image to an 8bit Image.
Keep in mind that converting an Image to 8bit will not necessary result in a loss of data. Check your Images, it might be that you are not using the full range of a 16bit value. you might be able to use a mixture of dynamic shifting and bit shifting in order to convert an image to an 8 bit, then embedding these criterias in the image and use them to convert back to a 16bit at a later time without losing any data, or in most cases minimal precision loss.
If you would like to attach one of the images you are using, I can take a look at it to see if this is possible in your case.
Good luck,
Dan,
