VISITRON: Visual Semantics-Aligned Interactively Trained Object-Navigator.
Ayush ShrivastavaKarthik GopalakrishnanYang LiuRobinson PiramuthuGökhan TürDevi ParikhDilek Hakkani-TurPublished in: ACL (Findings) (2022)
Keyphrases
- visual objects
- visual appearance
- visual representations
- visual properties
- spatial relations
- object model
- moving objects
- high level semantics
- semantically relevant
- visual input
- visual features
- complex objects
- real world objects
- similar objects
- logic programming
- object description
- low level
- neural network
- contextual cues
- formal semantics
- object models
- lighting conditions
- svm classifier
- semantic information
- d objects
- viewpoint
- three dimensional