What Should the System Do Next?: Operative Action Captioning for Estimating System Actions.
Taiki NakamuraSeiya KawanoAkishige YuguchiYasutomo KawanishiKoichiro YoshinoPublished in: CoRR (2022)
Keyphrases
- human actions
- reasoning about actions
- action sequences
- action selection
- recognizing actions
- action language
- concurrent actions
- action recognition
- action descriptions
- recognizing human actions
- joint action
- action detection
- action rules
- action theories
- action models
- partial observations
- situation calculus
- action primitives
- initial state
- indirect effects
- action space
- decision theoretic
- atomic actions
- continuous action
- state action
- body motions
- rigid body
- human activities
- activity recognition
- space time
- sensing actions
- image sequences
- speech acts
- plan execution
- state space
- spatio temporal
- state transitions
- multiple agents
- plan recognition
- visual features
- video sequences
- partially observable domains
- derived predicates
- neural network
- space time interest points