Robots learn by watching how-to videos

Scientists are teaching robots to watch how-to videos and derive a series of step-by-step instructions to perform a task, an advance that may help future 'personal robots' to do everyday housework such as cooking and washing dishes.
The researchers at the Cornell University in New York call their project "RoboWatch."
There is a common underlying structure to most how-to videos and there is plenty of source material available, researchers said.
YouTube offers 180,000 videos on "How to make an omelet" and 281,000 on "How to tie a bowtie."
By scanning multiple videos on the same task, a computer can find what they all have in common and reduce that to simple step-by-step instructions in natural language.

People post all these videos "to help people or maybe just to show off," said graduate student Ozan Sener, lead author of a paper on the method presented at the International Conference on Computer Vision in Chile.

The work is aimed at a future when we may have "personal robots" to perform everyday housework - cooking, washing dishes, doing the laundry, feeding the cat - as well as to assist the elderly and people with disabilities, researchers said.
A key feature of the system is that it is "unsupervised," said Sener who collaborated with colleagues at Stanford University, where he is currently a visiting researcher.

In most previous work, robot learning is accomplished by having a human explain what the robot is observing - for example, teaching a robot to recognise objects by showing it pictures of the objects while a human labels them by name.
In the new method, a robot with a job to do can look up the instructions and figure them out for itself.
Faced with an unfamiliar task, the robot's computer brain begins by sending a query to YouTube to find a collection of how-to videos on the topic.
The algorithm includes routines to omit "outliers" - videos that fit the keywords but are not instructional.

The computer scans the videos frame by frame, looking for objects that appear often, and reads the accompanying narration - using subtitles - looking for frequently repeated words.
Using these markers it matches similar segments in the various videos and orders them into a single sequence. From the subtitles of that sequence it can produce written instructions.
In other research, robots have learned to perform tasks by listening to verbal instructions from a human. In the future, information from other sources such as Wikipedia might be added.
The learned knowledge from the YouTube videos is made available via RoboBrain, an online knowledge base robots anywhere can consult to help them do their jobs.