Dense captioning algorithm applied to Boston Dynamics Atlas robots.
Originally made in 2016: https://vimeo.com/173025372
Densecap is an early image multi-captioning system developed at Stanford Vision Lab in 2015-16. This video was the result of an attempt to make it caption images outside of its training set, and poke fun at the lack of vocabulary in early AI img-to-text systems.
Press:
LaughingSquid: https://laughingsquid.com/dense-captioning-of-video-demonstrating-the-upgraded-boston-dynamics-atlas-robot/
Prosthetic Knowledge: https://prostheticknowledge.tumblr.com/post/146850706951/dense-captioning-of-boston-dynamics-atlas-robot
Original video from Boston Dynamics: https://www.youtube.com/watch?v=rVlhMGQgDkY
Densecap @ Stanford: https://cs.stanford.edu/people/karpathy/densecap
atayilgun
