Distilling Coarse-to-Fine Semantic Matching Knowledge for Weakly Supervised 3D Visual Grounding.
Zehan WangHaifeng HuangYang ZhaoLinjun LiXize ChengYichen ZhuAoxiong YinZhou ZhaoPublished in: ICCV (2023)
Keyphrases
- coarse to fine
- weakly supervised
- semantic matching
- multiresolution
- multiscale
- object detection
- object class
- image registration
- domain knowledge
- superpixels
- knowledge base
- dynamic programming
- relation extraction
- knowledge representation
- knowledge discovery
- ontology alignment
- computer vision
- learning algorithm
- web service composition
- semantic web services
- image processing
- named entities
- natural images
- machine learning
- semi supervised