Sampling Based Scene-Space Video Processing
Aug 07, 2015 by CGP Staff
0
|
Compelling video processing effects can be achieved if per-pixel depth information and 3D camera calibrations are known, such as denoising, stylized action shots, depth of field, scene-space masks, etc. However, this “scene-space” information is not always available.
Researchers from Disney Research Zurich and the Braunschweig University of Technology have come up with a new sampling-based framework for processing video that enables high-quality scene-space video effects in footage acquired casually. This includes hand-held, moving and uncalibrated video, along with potentially incorrect depth maps and camera pose information, both of which are either derived from the input footage itself or acquired using other sensors.
Watch a video on YouTube and find out more on Disney Research (PDF).
Researchers from Disney Research Zurich and the Braunschweig University of Technology have come up with a new sampling-based framework for processing video that enables high-quality scene-space video effects in footage acquired casually. This includes hand-held, moving and uncalibrated video, along with potentially incorrect depth maps and camera pose information, both of which are either derived from the input footage itself or acquired using other sensors.
Watch a video on YouTube and find out more on Disney Research (PDF).
