Ground Breaking Video Editing Technology

Posted on August 25, 2008 in Digital Effects | 8 Comments

The University of Washington who’s computer technology team recently developed Photosynth now has a group of graduate students showing the potential of technology to enhance digital video in amazing ways.

Video and Descriptions From: Pravin Bhat


Using Photographs to Enhance Videos of a Static Scene from pro on Vimeo.

The work presents a system for automatically producing a wide variety of video enhancements and visual effects. Unlike traditional visual effects software (e.g., After Effects, Shake, Boujou, etc), the system is completely automatic and no manual labor is required from the user. The major limitation of the work is that it can currently handle only videos of static scenes (i.e., videos shot with a moving camera but containing no moving objects in the scene). Efforts are being made to lift this restriction in future work.

Applications of the system include:
High resolution/definition video,
High dynamic range video,
Removing objects from a video,
Creating painterly (NPR) videos,
Video stabilization,
Easy video editing

Project website:
grail.cs.washington.edu/projects/videoenhancement/videoEnhancement.htm


Interactive Video Cutout from pro on Vimeo.

The work presents an interactive system for quickly creating alpha mattes for objects in a video. Unlike previous techniques commonly employed in the visual-fx industry (e.g., rotoscoping) our system can help novice users create professional quality mattes with very little manual effort.

These alpha mattes can then be used to cut out objects from one video and composite them into another video. Alpha mattes can also be used to apply image filters selectively to objects in a video.

Project website:
juew.org/projects/VideoCutout/VideoCutout.htm


GradientShop: A novel approach to image and video processing from pro on Vimeo.


Painterly rendering using GradientShop from pro on Vimeo.

The results shown in the video were created automatically by our painterly rendering filter. This filter was written using just 200 lines of C++ code by leveraging the GradientShop API.

To demonstrate the utility of our framework we have developed, or improved, several image processing applications, which include:
# Saliency sharpening
# Painterly rendering
# Deblocking
# Video deflickering
# Image relighting
# And sparse data interpolation (e.g., colorization)

You can see higher resolution results and get more information regarding GradientShop on the project website: GradientShop.com