The color contrast of all our picture styles is based on the original colorimetry we programmed for VisionColor but since we utilized different gamma curves for each of our profiles and, as lightness and saturation are co-related, we had to make some adjustments to individual color tones. Since the data being shifted towards the lower gamma range falls into the part of the image with the least data being assigned to by the compression, there is little to no poterization introduced in the midtones and highlights which has been our primary concern with all of our styles. In short, VisionTech raises the blacks in order to retain low light color information, especially in the red channel to preserve skin tones. Obviously this electronic boost results in an increase of noise but since our styles are tailored to a specific post-workflow, a little bit of noise should not be a concern and can easily be taken care of by Neat Video. Shifting the black-point away from 0% luminocity before encoding to 8-bit h.264 enables the camera to suck additional low-light color information out of the sensor.

It is the lowest 5% that are being boosted in VisionTech so that actual color information (H,S) is preserved where it would otherwise be lost due to luminocity values of 0-2. With our picture styles we have seperated the luma range into 20 increments of 5%. To explain, the internal processing of the picture styles employ the HSL standard which stands for Hue, Saturation and Luminocity. (3% or RGB 9,9,9)ĭespite popular belief this is not because it gives more latitude (which it doesn't) but to enable the camera to encode color information where in other profiles there would be none. VisionColor utilizes the entire tonal range available in the 8-bit coding space, clipping at 0 and 100% luminocity (RGB values of 0,0,0 and 255,255,255) whereas VisionTech performs a slight boost of the blackpoint.

Canon picture style john hope cinema