From time to time I read an article on the Internet about a new technology or software that makes me jump up and down and clap my hands at the same time.
In other words, a technology that blows me away. This time, it is a new method for enhancing video by using photos of the same scene to create effects like high definition resolutions, enhanced exposure, object removal, object touchup or high dynamic range.
Computer scientists at the University of Washington have developed a software application that automatically enhances a video by using several photographs of the same scene.
This takes the higher resolution and detail of photos in comparison to videos into account to create stunning videos.
Take a look at the original images in the upper row and the enhanced images in the lower row.
The photos highlight the different options available to users of the program. From left to right those are:
While it might be nice to see the differences on photos, it is more interesting to see a video with moving pictures and a voice explaining the technique in greater detail.
The technique has a few limitations. It currently works only on static scenes, meaning scenes with no moving objects except for the camera and lighting. It furthermore takes roughly 5 minutes to compute one 800x600 frame.
A code dump of the project is available on the project's website which might be useful as a reference for implementing the algorithm.
You can also download the paper in pdf format, the source code, and talks (slides with videos or only slides) that provide you with additional details about the technology.Advertisement
Ghacks is a technology news blog that was founded in 2005 by Martin Brinkmann. It has since then become one of the most popular tech news sites on the Internet with five authors and regular contributions from freelance writers.