Which One? Leveraging Context Between Objects and Multiple Views for Language Grounding.
Chancharik MitraAbrar AnwarRodolfo CoronaDan KleinTrevor DarrellJesse ThomasonPublished in: NAACL-HLT (2024)
Keyphrases
- multiple views
- d objects
- multi view
- viewpoint
- single view
- uncalibrated images
- multiple viewpoints
- camera views
- word meanings
- point correspondences
- dynamic scenes
- multiple cameras
- multiple images
- consistency constraints
- three dimensional
- fundamental matrix
- ground plane
- planar surfaces
- multiple objects
- point features
- range images
- homography estimation
- overlapping views
- multiple range images
- geometric constraints
- range data
- computer vision