Make3D Converts your Photo into a 3-D Model
Make3D is a web service that turns your single picture into a three-dimensional “fly around” model. This will give you a new way of looking at your photos. It will allow you to access a scene’s depth and a range of points of view. Check out the video below.
The service is free and you can upload a photo from your hard drive or Flickr account. All you have to do in order to upload a photo is to sign up for an account. After uploading your image/photo you can view, or in their words “fly” in the 3-D scene. To do this, you must use a VRML (Virtual Reality Modeling Language) viewer or Adobe Shockwave. A rendered 3-D flash movie is also made available for you to watch if you do not have or don’t want to go to the trouble of downloading and installing the above mentioned software. You can also view your created 3-D model in high resolution movie in 3DS format (this feature is still unavailable at the moment). The fun doesn’t end in viewing your photos in 3-D. Make3D provides a code so you can embed the 3-D model in your websites and social networking pages.I have to warn you though that once you have uploaded your image, it takes some time for the program to create the 3-D model. The project is still in research stage and they are only using one computer for this which is most of the time is overloaded. But, as soon as your image is queued, you will be able to watch and download your 3-D model.
How it works:
Our software uses a breakthrough technology in machine learning. It estimates depths from the single image by using our monocular vision algorithm, developed in 2005. It captures a variety of monocular cues and learns the relations between different parts of the image using a machine learning technique called Markov Random Field (MRF). Our algorithm first divides the image into small patches and analyzes them at multiple scales to estimate each of the patches’ 3-d location and 3-d orientation.
We have applied variations of this algorithm for driving cars autonomously, for robotic manipulation, for making 3-d models of large environments, and for creating visually-pleasing 3-d throughs from an image.
For hard-core programmers, you can download the codes for the program that they are using here.