Given a still image, CSAIL deep-learning system generates videos that predict what will happen next in a scene.
Living in a dynamic physical world, it’s easy to forget how effortlessly we understand our surroundings. With minimal thought, we can figure out how scenes change and objects interact.
But what’s second nature for us is still a huge problem for machines. With the limitless number of ways that objects can move, teaching computers to predict future actions can be difficult.
Recently, researchers from MIT’s Computer Science and Artificial Intelligence Laboratory(CSAIL) have moved a step closer, developing a deep-learning algorithm that, given a still image from a scene, can create a brief video that simulates the future of that scene.
Trained on 2 million unlabeled videos that include a year’s worth of footage, the algorithm generated videos that human subjects deemed to be realistic 20 percent more often than a baseline model.
The team says that future versions could be used for everything from improved security tactics and safer self-driving cars. According to CSAIL PhD student and first author Carl Vondrick, the algorithm can also help machines recognize people’s activities without expensive human annotations.
“These videos show us what computers think can happen in a scene,” says Vondrick. “If you can predict the future, you must have understood something about the present.”
Vondrick wrote the paper with MIT professor Antonio Torralba and Hamed Pirsiavash, a former CSAIL postdoc who is now a professor at the University of Maryland Baltimore County (UMBC). The work will be presented at next week’s Neural Information Processing Systems (NIPS) conference in Barcelona.
How it works
Multiple researchers have tackled similar topics in computer vision, including MIT Professor Bill Freeman, whose new work on “visual dynamics” also creates future frames in a scene. But where his model focuses on extrapolating videos into the future, Torralba’s model can also generate completely new videos that haven’t been seen before.
Previous systems build up scenes frame by frame, which creates a large margin for error. In contrast, this work focuses on processing the entire scene at once, with the algorithm generating as many as 32 frames from scratch per second.
“Building up a scene frame-by-frame is like a big game of ‘Telephone,’ which means that the message falls apart by the time you go around the whole room,” says Vondrick. “By instead trying to predict all frames simultaneously, it’s as if you’re talking to everyone in the room at once.”
Of course, there’s a trade-off to generating all frames simultaneously: While it becomes more accurate, the computer model also becomes more complex for longer videos. Nevertheless, this complexity may be worth it for sharper predictions.
To create multiple frames, researchers taught the model to generate the foreground separate from the background, and to then place the objects in the scene to let the model learn which objects move and which objects don’t.
The team used a deep-learning method called “adversarial learning” that involves training two competing neural networks. One network generates video, and the other discriminates between the real and generated videos. Over time, the generator learns to fool the discriminator.
From that, the model can create videos resembling scenes from beaches, train stations, hospitals, and golf courses. For example, the beach model produces beaches with crashing waves, and the golf model has people walking on grass.
Testing the scene
The team compared the videos against a baseline of generated videos and asked subjects which they thought were more realistic. From over 13,000 opinions of 150 users, subjects chose the generative model videos 20 percent more often than the baseline.
Vondrick stresses that the model still lacks some fairly simple common-sense principles. For example, it often doesn’t understand that objects are still there when they move, like when a train passes through a scene. The model also tends to make humans and objects look much larger in size than reality.
Another limitation is that the generated videos are just one and a half seconds long, which the team hopes to be able to increase in future work. The challenge is that this requires tracking longer dependencies to ensure that the scene still makes sense over longer time periods. One way to do this would be to add human supervision.
“It’s difficult to aggregate accurate information across long time periods in videos,” says Vondrick. “If the video has both cooking and eating activities, you have to be able to link those two together to make sense of the scene.”
These types of models aren’t limited to predicting the future. Generative videos can be used for adding animation to still images, like the animated newspaper from the Harry Potter books. They could also help detect anomalies in security footage and compress data for storing and sending longer videos.
“In the future, this will let us scale up vision systems to recognize objects and scenes without any supervision, simply by training them on video,” says Vondrick.
Learn more: Creating videos of the future
The Latest on: Predictive artificial intelligence
via Google News
The Latest on: Predictive artificial intelligence
- AAAI 2021: Accelerating the impact of artificial intelligenceon February 24, 2021 at 4:43 pm
The purpose of the Association for the Advancement of Artificial Intelligence, according to its bylaws, is twofold. The first is to promote research in the area of AI, and the second is to promote the ...
- METIS first with predictive Poseidon Principles Emissions INDEXon February 24, 2021 at 2:49 pm
METIS Cyberspace Technology has refined its cloud-based data acquisition and ship performance reporting solution to offer shipping’s first tool to predict the trade-off between emissions reduction and ...
- SAIC names senior VP of artificial intelligenceon February 24, 2021 at 1:19 pm
Reston-based Fortune 500 defense contractor Science Applications International Corp. (SAIC) has hired Michael Scruggs as senior vice president of artificial intelligence, effective immediately. “In ...
- Centro Acquires AI-Driven Search Advertising and Marketing Intelligence Platform QuanticMindon February 24, 2021 at 9:30 am
Shawn Riegsecker, CEO of Centro “Centro is far ahead in building for marketers the most comprehensive omnichannel digital advertising platform. QuanticMind drives our vision by enabling us to weave ...
- Data Science and Artificial Intelligence Is Revolutionizing The Sports Industryon February 23, 2021 at 1:11 pm
Data science in the sports industry offers sports data analytics. AI and machine learning models can help evaluate individual player performance, team performance, and help teams make decisions that ...
- Lone Star Analysis Launches MaxUp™, a Real-Time Predictive and Prescriptive Asset Analytics Software Suiteon February 23, 2021 at 6:50 am
Lone Star Analysis, a trusted provider of leading-edge predictive and prescriptive analytics, and guided artificial intelligence solutions, announced today the launch of its real-time ...
- Forecasting the evolution of artificial intelligence: meeting with Emile Servan-Schreiberon February 22, 2021 at 4:19 pm
On the occasion of the contest to predict the evolution of artificial intelligence launched by Hypermind with the support of Open Philanthropy, we decided to learn more about the initiative from Emile ...
- Take A Look At How Artificial Intelligence Can Improve Your Pet’s Healthon February 22, 2021 at 3:13 pm
Could artificial intelligence give us better insight into our pet's health? How can a data-driven approach help animals live healther, happier lives?
- Explainable Artificial Intelligence for Decoding Regulatory Instructions in DNAon February 21, 2021 at 5:41 am
Opening the black box to uncover the rules of the genome’s regulatory code. Researchers at the Stowers Institute for Medical Research, in collaboration with colleagues at Stanford University and ...
via Bing News