TIGEr: Text-to-Image Grounding for Image Caption Evaluation.
Ming JiangQiuyuan HuangLei ZhangXin WangPengchuan ZhangZhe GanJana DiesnerJianfeng GaoPublished in: EMNLP/IJCNLP (1) (2019)
Keyphrases
- image features
- single image
- image data
- multiscale
- image content
- input image
- image classification
- template matching
- image representation
- image pixels
- image retrieval
- image analysis
- image structure
- segmentation method
- image collections
- lighting conditions
- high resolution
- caption text
- vector field
- image regions
- low level
- feature points
- edge detection
- image matching
- similarity measure
- pixel values
- medical images