Using ultrafine-grained Graph-RISE embeddings instead of Faster R-CNN features, with the same proposed boxes, improves image captioning on Conceptual Captions and VQA on VizWiz.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Decoupled Box Proposal and Featurization with Ultrafine-Grained Semantic Labels Improve Image Captioning and Visual Question Answering
Using ultrafine-grained Graph-RISE embeddings instead of Faster R-CNN features, with the same proposed boxes, improves image captioning on Conceptual Captions and VQA on VizWiz.