Add to Favourites
To login click here
Researchers at Carnegie Mellon University’s Robotics Institute have developed a framework called MoTok that enables a computer to identify features of objects in motion and use those features to reconstruct the object. This could help computers and robots better automatically detect objects in videos, which could be useful for autonomous driving, retail robotics, robotic manipulation and robots in the home. The research team has also extended the work so the computer can depict these features in a simplified, virtualized fashion, allowing the computer to better identify high-level features and categorize objects.
