Machine learning system efficiently recognizes activities by observing how objects change in only a few key frames.
Given only a few frames of a video, humans can usually surmise what is happening and will happen on screen. If we see an early frame of stacked cans, a middle frame with a finger at the stack’s base, and a late frame showing the cans toppled over, we can guess that the finger knocked down the cans. Computers, however, struggle with this concept.
In a paper being presented at this week’s European Conference on Computer Vision, MIT researchers describe an add-on module that helps artificial intelligence systems called convolutional neural networks, or CNNs, to fill in the gaps between video frames to greatly improve the network’s activity recognition.
The researchers’ module, called Temporal Relation Network (TRN), learns how objects change in a video at different times. It does so by analyzing a few key frames depicting an activity at different stages of the video — such as stacked objects that are then knocked down. Using the same process, it can then recognize the same type of activity in a new video.
In experiments, the module outperformed existing models by a large margin in recognizing hundreds of basic activities, such as poking objects to make them fall, tossing something in the air, and giving a thumbs-up. It also more accurately predicted what will happen next in a video — showing, for example, two hands making a small tear in a sheet of paper — given only a small number of early frames.
One day, the module could be used to help robots better understand what’s going on around them.
“We built an artificial intelligence system to recognize the transformation of objects, rather than appearance of objects,” says Bolei Zhou, a former PhD student in the Computer Science and Artificial Intelligence Laboratory (CSAIL) who is now an assistant professor of computer science at the Chinese University of Hong Kong. “The system doesn’t go through all the frames — it picks up key frames and, using the temporal relation of frames, recognize what’s going on. That improves the efficiency of the system and makes it run in real-time accurately.”
Co-authors on the paper are CSAIL principal investigator Antonio Torralba, who is also a professor in the Department of Electrical Engineering and Computer Science; CSAIL Principal Research Scientist Aude Oliva; and CSAIL Research Assistant Alex Andonian.
Picking up key frames
Two common CNN modules being used for activity recognition today suffer from efficiency and accuracy drawbacks. One model is accurate but must analyze each video frame before making a prediction, which is computationally expensive and slow. The other type, called two-stream network, is less accurate but more efficient. It uses one stream to extract features of one video frame, and then merges the results with “optical flows,” a stream of extracted information about the movement of each pixel. Optical flows are also computationally expensive to extract, so the model still isn’t that efficient.
“We wanted something that works in between those two models — getting efficiency and accuracy,” Zhou says.
The researchers trained and tested their module on three crowdsourced datasets of short videos of various performed activities. The first dataset, called Something-Something, built by the company TwentyBN, has more than 200,000 videos in 174 action categories, such as poking an object so it falls over or lifting an object. The second dataset, Jester, contains nearly 150,000 videos with 27 different hand gestures, such as giving a thumbs-up or swiping left. The third, Charades, built by Carnegie Mellon University researchers, has nearly 10,000 videos of 157 categorized activities, such as carrying a bike or playing basketball.
When given a video file, the researchers’ module simultaneously processes ordered frames — in groups of two, three, and four — spaced some time apart. Then it quickly assigns a probability that the object’s transformation across those frames matches a specific activity class. For instance, if it processes two frames, where the later frame shows an object at the bottom of the screen and the earlier shows the object at the top, it will assign a high probability to the activity class, “moving object down.” If a third frame shows the object in the middle of the screen, that probability increases even more, and so on. From this, it learns object-transformation features in frames that most represent a certain class of activity.
Recognizing and forecasting activities
In testing, a CNN equipped with the new module accurately recognized many activities using two frames, but the accuracy increased by sampling more frames. For Jester, the module achieved top accuracy of 95 percent in activity recognition, beating out several existing models.
It even guessed right on ambiguous classifications: Something-Something, for instance, included actions such as “pretending to open a book” versus “opening a book.” To discern between the two, the module just sampled a few more key frames, which revealed, for instance, a hand near a book in an early frame, then on the book, then moved away from the book in a later frame.
Some other activity-recognition models also process key frames but don’t consider temporal relationships in frames, which reduces their accuracy. The researchers report that their TRN module nearly doubles in accuracy over those key-frame models in certain tests.
The module also outperformed models on forecasting an activity, given limited frames. After processing the first 25 percent of frames, the module achieved accuracy several percentage points higher than a baseline model. With 50 percent of the frames, it achieved 10 to 40 percent higher accuracy. Examples include determining that a paper would be torn just a little, based how two hands are positioned on the paper in early frames, and predicting that a raised hand, shown facing forward, would swipe down.
“That’s important for robotics applications,” Zhou says. “You want [a robot] to anticipate and forecast what will happen early on, when you do a specific action.”
Next, the researchers aim to improve the module’s sophistication. The first step is implementing object recognition together with activity recognition. Then, they hope to add in “intuitive physics,” meaning helping it understand real-world physical properties of objects. “Because we know a lot of the physics inside these videos, we can train module to learn such physics laws and use those in recognizing new videos,” Zhou says. “We also open source all the code and models. Activity understanding is an exciting area of artificial intelligence right now.”
The Latest on: Machine learning
via Google News
The Latest on: Machine learning
- Machine learning on Raspberry Pi just took a big step forwardon April 14, 2021 at 1:58 am
Raspberry Pi is a capable little machine, but if you're interested in developing your own embedded machine-learning applications, training custom models on the platform has historically been tricky ...
- New Report: Behavioral Analytics, Machine Learning Keep FIs On Cutting Edge Of Fraud Preventionon April 14, 2021 at 1:07 am
The Financial Fraud Prevention Playbook examines the fraud threats facing FIs and how banks can use machine learning and behavioral analytics to fight back.
- Machine learning can help keep the global supply chain movingon April 13, 2021 at 9:54 am
Disruptions can come from almost anywhere, and keeping goods moving is critical. One expert says AI and machine learning can help.
- Machine learning, transcriptome, and genotyping chip analyses provide insights into SNP markers identifying flower color in Platycodon grandifloruson April 13, 2021 at 9:02 am
Bellflower is an edible ornamental gardening plant in Asia. For predicting the flower color in bellflower plants, a transcriptome-wide approach based on machine learning, transcriptome, and genotyping ...
- Open-Source Framework Simplifies Machine Learning Processeson April 13, 2021 at 6:31 am
April 13, 2021 - Researchers at MIT’s Data to AI Lab (DAI Lab) have developed a new framework that can streamline machine learning processes to help organizations uncover actionable insights from big ...
- Machine learning can help slow down future pandemicson April 13, 2021 at 4:55 am
Artificial intelligence could be one of the keys for limiting the spread of infection in future pandemics. In a new study, researchers at the University of Gothenburg have investigated how machine ...
- ANZ Bank: We've been using machine learning for 20 yearson April 12, 2021 at 11:17 pm
Although the concept of machine learning isn't new to ANZ, it says it now possesses the compute power to move full-steam ahead.
- How Machine Learning Will Enable Technologies That Anticipate What The Brain Thinkson April 12, 2021 at 9:56 pm
And now, converging with advances taking place in machine learning, BMIs are on the verge of entirely new capabilities. The Problem with One-Size-Fits-All BMIs There is a tremendous amount of ...
- One-stop machine learning platform turns health care data into insightson April 12, 2021 at 8:01 am
Over the past decade, hospitals and other health care providers have put massive amounts of time and energy into adopting electronic health care records, turning hastily scribbled doctors' notes into ...
via Bing News