Website review: Scene Completion Using Millions of ...
mmmPi discovered this in Photography
•9 reviews since Jul 10, 2007
photography
•graphics.cs.cmu.edu/projects/scene-completion...
People who like this website
- Linchung
San Diego
- Donnie373
Lompoc
- geneuine
California
- hobobot
Napa
- swdunn
Orland
- mdavidl
Tucson
- valerionson
Orem
- ifsck
Salt Lake City
- jinxdog
Oregon City
- bryanhansen
Colorado
StumbleUpon is the best way to discover great web sites, videos, photos, blogs and more - based on your interests.
Everything is submitted and rated by the community. Discover, share and review the best of the web!
Reviews of this website
mmmPi discovered 13 months ago- Cool but creepy.
- famine-weasel rated 10 months ago
- Cool.
- mrector rated 10 months ago
- Interesting, just as photosynth is.
- onomatoh rated 11 months ago
- Fill in the missing parts using millions of images. Unbelievable.
- KevinFairchild rated 11 months ago
- This looks cool -- assuming it works as well as it's being advertised...
- eliasen rated 12 months ago
- Very impressive. The paper presents algorithms that "patch up" missing parts of images with other images from a large database. Look at the pictures in the .pdf file if you do nothing else. I want to fiddle with this software! Since it's supported by NSF grants, one should be able to file a freedom of information request and get the software...
- norteo rated 12 months ago
- From the page: "What can you do with a million images? In this paper we present a new image completion algorithm powered by a huge database of photographs gathered from the Web. The algorithm patches up holes in images by finding similar image regions in the database that are not only seamless but also semantically valid. Our chief insight is that while the space of images is effectively infinite, the space of semantically differentiable scenes is actually not that large. For many image completion tasks we are able to find similar scenes which contain image fragments that will convincingly complete the image. Our algorithm is entirely data-driven, requiring no annotations or labelling by the user. Unlike existing image completion methods, our algorithm can generate a diverse set of image completions and we allow users to select among them. We demonstrate the superiority of our algorithm over existing image completion approaches."
- tranquil222 rated 12 months ago
- Interesting but where's a useful real life application?
- ultral rated 13 months ago
- These guy are god of algoritmic !
