HIGHLIGHTS
- who: Sen Du and colleagues from the School of Automation and Information Engineering, Xi`an University of Technology, Xi`an, China have published the article: Controllable Image Captioning with Feature Refinement and Multilayer Fusion, in the Journal: (JOURNAL)
- what: This work proposes a novel module called the information-augmented graph encoder, which is composed of an information-augmented embedding module and a multi-relational weighted graph encoder to incorporate a priori information into objects or relation nodes, improving the accuracy and diversity of the generated sentences. In the future, the authors propose to design an . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.