Facebook Collected
More Than 2,200 Hours, of First-Person
Video To Train AI.
Facebook Collected
More Than 2,200 Hours, of First-Person
Video To Train AI.
The research project was announced by the lead scientist of Facebook's Reality Labs division on Oct. 14. .
Known as Ego4D, the project collected video captured from a first-person human perspective through devices like GoPro cameras and smart glasses.
The collection of footage could be used to teach AI to recognize and understand elements of the real world.
Researchers outside of Facebook will have access to the data beginning next month.
This release, which is an open data set and research challenge, is going to catalyze progress for us internally... , Kristen Grauman, Facebook's Lead Research Scientist, via CNBC.
... but also widely externally in the academic community and [allow] other researchers to get behind these
new problems .., Kristen Grauman, Facebook's Lead Research Scientist, via CNBC.
... but now be able to do it in a more meaningful way and at a greater scale, Kristen Grauman, Facebook's Lead Research Scientist, via CNBC.
According to Facebook leaders, the video data could be used to train the next generation of robots.
Traditionally a robot learns by doing stuff in the world or being literally handheld to be shown how to do things, Kristen Grauman, Facebook's Lead Research Scientist, via CNBC.
There’s openings to let them learn from video just from our own experience, Kristen Grauman, Facebook's Lead Research Scientist, via CNBC.
More than 700 participants from nine different countries captured the footage.
For this project, Facebook has partnered with universities all over the world.
An important design decision for this project is we wanted partners that first of all are leading experts in the field, .., Kristen Grauman, Facebook's Lead Research Scientist, via CNBC.
... interested in these problems and motivated to pursue them but also have geographic diversity, Kristen Grauman, Facebook's Lead Research Scientist, via CNBC