Vision-based method for semantic information extraction in construction by integrating deep learning object detection and image captioning.
Yiheng WangBo XiaoAhmed BoufergueneMohamed Al-HusseinHeng LiPublished in: Adv. Eng. Informatics (2022)
Keyphrases
- segmentation method
- information extraction
- deep learning
- object detection
- similarity measure
- image regions
- machine learning
- test images
- energy function
- unsupervised learning
- pairwise
- region of interest
- multiscale
- image segmentation
- higher order
- maximum likelihood
- text mining
- natural language
- image content
- named entity recognition
- data mining