I found that nice short movie, 1 on YouTube. It looks real world because I felt that I was inside the room. But it is a kind of artificial world created based on a few real images. This technology is a little complicated. In short, we can create a 3D world from 2D data as shown in the movie. It is amazing!
According to this paper, 2 this movie created by an artificial intelligence by Google. This system is called “Deep stereo”. When we look at Google street view, we do not feel this looks real because these images are limited by angles where they are taken. In order to make it look real, deep stereo synthesizes original images which are taken from several different angles and fill out the missing parts. It is said that “To our knowledge, our work is the first to apply deep learning to the problem of new view synthesis from sets of real world, natural imagery.” in the paper.
The figure above is taken from the paper (p3). When we have pictures taken from angles of V1 and V2, we can synthesize them to create the image seen from position C. It needs a lot of computations so it takes time to be completed. It should be improved in terms of speed if this technology is used in a real-time basis, according to the paper.
It means that when we have a few photos about one object, we might create our artificial world which looks so real. For example, when you have your own photos in new fashion, you could create your promotional video as a 3D movie. We could consider many applications. In restaurants, we could have electronic menus where each dish looks so real. For travel industries, they could present courses of journeys as if we could experience unknown real world. For real estate industries, they could provide the experience inside houses to potential buyers before it starts being constructed.
Therefore, this technology has huge potentials. Artificial intelligence is currently used for machine translation, face recognition, autonomous-driving and so on. This synthesized world creation is also expected to develop rapidly because big IT companies, such as Google, Facebook, Microsoft are investing a lot money in technologies of Virtual Reality and Augmented Reality. These are expected to make huge markets in the future.
Now many people enjoy taking their own pictures by themselves. We might create a 3D world based on these pictures in the future. We could zoom in and out with angles, which are different from angles where pictures were taken. So everyone could create a promotional video for him/herself. Is it fun, isn’t it?
1. Google’s ‘Deep Stereo’ could transform Street View into an immersive digital world, YouTube, 7 July 2015
2. DeepStereo: Learning to Predict New Views from the World’s Imagery, arXiv:1506.06825, 22 June 2015
Note: Toshifumi Kuga’s opinions and analyses are personal views and are intended to be for informational purposes and general interest only and should not be construed as individual investment advice or solicitation to buy, sell or hold any security or to adopt any investment strategy. The information in this article is rendered as at publication date and may change without notice and it is not intended as a complete analysis of every material fact regarding any country, region market or investment.
Data from third-party sources may have been used in the preparation of this material and I, Author of the article has not independently verified, validated such data. I accept no liability whatsoever for any loss arising from the use of this information and relies upon the comments, opinions and analyses in the material is at the sole discretion of the user.