Spoiler alert_ artificial intelligence can predict how scenes will play out – scientific american

A new artificial intelligence system can take still images and generate short videos that simulate what happens next similar to how humans can visually imagine how a scene will evolve, according to a new study.

Humans intuitively understand how the world works, which makes it easier for people, as opposed to machines, to envision how a scene will play out. Best beach resorts in barcelona spain But objects in a still image could move and interact in a multitude of different ways, making it very hard for machines to accomplish this feat, the researchers said. Beaches in barcelona in june But a new, so-called deep-learning system was able to trick humans 20 per cent of the time when compared to real footage.

Researchers at the Massachusetts Institute of Technology (MIT) pitted two neural networks against each other, with one trying to distinguish real videos from machine-generated ones, and the other trying to create videos that were realistic enough to trick the first system.


Beach resorts near to barcelona [ Super-Intelligent Machines: 7 Robotic Futures]

This kind of setup is known as a “generative adversarial network” (GAN), and competition between the systems results in increasingly realistic videos. Best beach barcelona tripadvisor When the researchers asked workers on Amazon’s Mechanical Turk crowdsourcing platform to pick which videos were real, the users picked the machine-generated videos over genuine ones 20 percent of the time, the researchers said. Good beaches in barcelona Early stages

Still, budding film directors probably don’t need to be too concerned about machines taking over their jobs yet—the videos were only 1 to 1.5 seconds long and were made at a resolution of 64 x 64 pixels. Best beach restaurants barcelona But the researchers said that the approach could eventually help robots and self-driving cars navigate dynamic environments and interact with humans, or let Facebook automatically tag videos with labels describing what is happening.

“Our algorithm can generate a reasonably realistic video of what it thinks the future will look like, which shows that it understands at some level what is happening in the present,” said Carl Vondrick, a Ph.D. Beaches in barcelona photos student in MIT’s Computer Science and Artificial Intelligence Laboratory, who led the research. Barcelona beaches map “Our work is an encouraging development in suggesting that computer scientists can imbue machines with much more advanced situational understanding.”

The system is also able to learn unsupervised, the researchers said. Best beach hotels near barcelona This means that the two million videos—equivalent to about a year’s worth of footage—that the system was trained on did not have to be labeled by a human, which dramatically reduces development time and makes it adaptable to new data.

In a study that is due to be presented at the Neural Information Processing Systems (NIPS) conference, which is being held from Dec. Top 10 beaches in barcelona 5 to 10 in Barcelona, Spain, the researchers explain how they trained the system using videos of beaches, train stations, hospitals and golf courses.

“In early prototypes, one challenge we discovered was that the model would predict that the background would warp and deform,” Vondrick told Live Science. Beaches near barcelona airport To overcome this, they tweaked the design so that the system learned separate models for a static background and moving foreground before combining them to produce the video. Map of barcelona beaches AI filmmakers

The MIT team is not the first to attempt to use artificial intelligence to generate video from scratch. Beach resort near barcelona el prat airport But, previous approaches have tended to build video up frame by frame, the researchers said, which allows errors to accumulate at each stage. Barcelona spain beaches photos Instead, the new method processes the entire scene at once—normally 32 frames in one go.

Ian Goodfellow, a research scientist at the nonprofit organization OpenAI, who invented GAN, said that systems doing earlier work in this field were not able to generate both sharp images and motion the way this approach does. Beach hotels near barcelona spain However, he added that a new approach that was unveiled by Google’s DeepMind AI research unit last month, called Video Pixel Networks (VPN), is able to produce both sharp images and motion. Beach barcelona hotels [ The 6 Strangest Robots Ever Created]

“Compared to GANs, VPN are easier to train, but take much longer to generate a video,” he told Live Science. Barcelona spain beaches pictures “VPN must generate the video one pixel at a time, while GANs can generate many pixels simultaneously.”

Vondrick also points out that their approach works on more challenging data like videos scraped from the web, whereas VPN was demonstrated on specially designed benchmark training sets of videos depicting bouncing digits or robot arms.

The results are far from perfect, though. Beaches near barcelona el prat Often, objects in the foreground appear larger than they should, and humans can appear in the footage as blurry blobs, the researchers said. Spain barcelona beaches Objects can also disappear from a scene and others can appear out of nowhere, they added.

“The computer model starts off knowing nothing about the world. Beaches in barcelona tripadvisor It has to learn what people look like, how objects move and what might happen,” Vondrick said. Best beaches barcelona map “The model hasn’t completely learned these things yet. Quiet beaches near barcelona Expanding its ability to understand high-level concepts like objects will dramatically improve the generations.”

Another big challenge moving forward will be to create longer videos, because that will require the system to track more relationships between objects in the scene and for a longer time, according to Vondrick.

“To overcome this, it might be good to add human input to help the system understand elements of the scene that would be difficult for it to learn on its own,” he said.