CgT-GAN: CLIP-guided Text GAN for Image Captioning.
Jiarui YuHaoran LiYanbin HaoBin ZhuTong XuXiangnan HePublished in: ACM Multimedia (2023)
Keyphrases
- image data
- input image
- image features
- single image
- image segmentation
- multiscale
- low level
- image content
- image classification
- image analysis
- text information
- test images
- image matching
- hough transform
- image regions
- image representation
- text mining
- high resolution
- gray scale
- keypoints
- similarity measure
- template matching
- structuring elements
- information retrieval
- computer vision
- region of interest
- image set
- textual information
- web images