MLIN, a stacked attention-based network that reasons over latent summarizations of image regions and question words, achieves competitive VQA v2.0 and TDIUC accuracy with reduced message-passing cost.
Bottom-up and top-down attention for image captioning and visual question answering
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Multi-modality Latent Interaction Network for Visual Question Answering
MLIN, a stacked attention-based network that reasons over latent summarizations of image regions and question words, achieves competitive VQA v2.0 and TDIUC accuracy with reduced message-passing cost.