Open realizator opened 5 years ago
As intended. You have created a stereo pair of cameras. All operations will apply equally to both image processing paths. The results of those two paths are then combined into a single output frame.
Ok, thank you. This issue can be closed now.
I tried to use "zoom" (ex. "crop") in stereoscopic side-by-side mode, suggesting this option will be applied to captured stereoscopic image. My aim was to take left and right images from stereopair using GPU-accelerated crop. But in fact all settings implemented to both sensors, and after that final stereoscopic image is constructed by stitching this images. So instead of cropped left part of stereoscopic image I obtain stereoscopic image from two left parts of images captured by cameras. Is this behavior was originally planned, or was just not tested in stereoscopic mode?