Extracting Zero-shot Common Sense from Large Language Models for Robot 3D Scene Understanding.
William ChenSiyi HuRajat TalakLuca CarlonePublished in: CoRR (2022)
Keyphrases
- d scene
- language model
- language modeling
- single image
- probabilistic model
- depth map
- document retrieval
- retrieval model
- language modelling
- n gram
- information retrieval
- speech recognition
- scene understanding
- statistical language models
- smoothing methods
- optical flow
- pseudo relevance feedback
- test collection
- language models for information retrieval
- vision system
- query expansion
- context sensitive
- relevance model
- camera viewpoint
- scene flow
- data sets
- multi view
- d objects
- document ranking
- flow estimation