Unifying Multimodal Transformer for Bi-directional Image and Text Generation.
Yupan HuangHongwei XueBei LiuYutong LuPublished in: CoRR (2021)
Keyphrases
- bi directional
- input image
- image classification
- text generation
- single image
- image features
- multiscale
- image retrieval
- image analysis
- feature points
- image content
- image data
- image representation
- image segmentation
- test images
- high resolution
- segmentation method
- multi modal
- natural language generation
- artificial intelligence
- edge detection
- segmentation algorithm
- probability distribution