TIGEr: Text-to-Image Grounding for Image Caption Evaluation.
Ming JiangQiuyuan HuangLei ZhangXin WangPengchuan ZhangZhe GanJana DiesnerJianfeng GaoPublished in: CoRR (2019)
Keyphrases
- template matching
- input image
- image content
- single image
- image data
- image segmentation
- image features
- multiscale
- segmentation method
- image representation
- low level
- image classification
- edge detection
- web images
- lighting conditions
- test images
- hough transform
- feature vectors
- image regions
- high resolution
- image analysis
- image matching
- image retrieval
- vector field
- region of interest
- video sequences
- scale space
- image collections
- image structure
- image pixels
- bounding box
- edge map
- image quality