Google developed its own video stabilization technology and algorithm for this and is now presenting it at IEEE's International Conference on Computer Vision and Pattern Recognition.
It's also making it available online to the benefit of other researchers and professionals working on video technologies.
"Existing in-camera stabilization methods dampen high-frequency jitter but do not suppress low-frequency movements and bounces, such as those observed in videos captured by a walking person," Matthias Grundmann, Vivek Kwatra and Irfan Essa, from Google Research wrote.
"Our goal was to devise a completely automatic method for converting casual shaky footage into more pleasant and professional looking videos," they added.
They had to come up with an algorithm that not only found the best 'path' for the camera to take to make the image look stable without losing too much detail, since the stabilization algorithm essentially relies on cropping the individual frames, but also one that was efficient enough to enable real-time editing.
"Our technique mimics the cinematographic principles outlined above by automatically determining the best camera path using a robust optimization technique," they explained.
"The original, shaky camera path is divided into a set of segments, each approximated by either a constant, linear or parabolic motion. Our optimization finds the best of all possible partitions using a computationally efficient and stable algorithm," they added.
No comments:
Post a Comment