Action prediction based on physically grounded object affordances in human-object interactions.

Vibekananda Dutta Teresa Zielinska

Published in: RoMoCo (2017)

Keyphrases

human object interactions
human actions
action recognition
human pose
complex events
spatio temporal
d objects
space time
human motion
video sequences
object tracking
human body
pose estimation
data processing
machine learning
bounding box
computer vision