Convert 2D Images into 3D Images

Make 3D is a wonderful piece of code from the University of Stanford that transforms a normal 2D image into a 3D image. While the results are not true 3D images they are still very impressive. I tested the script with two images that I took last summer, one from the North Sea [link] and one from my trip to Berlin showing Berlin Cathedral [link].

No user input at all other than selecting an image from Flickr or your computer and tagging it is required. The script is transforming the image automatically one after the other. You receive a message about your queue position and will receive an email once the transformation finishes.

Users can then zoom in and out of the image, turn, move forward, backward and up and down using the cursor keys. Adobe Shockwave is required for the script to work.

How does it work ?

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. More details could be found here.

Feel free to share your images. The website of the service lists user generated images on its pages that you can browse to get an impression of what the technology has to offer. Note that you need to have Shockwave installed on your system to view the sample images on the website.

Advertisement