Neural Module Networks

Visual question answering is fundamentally compositional in nature-a question like where is the dog? shares substructure with questions like what color is the dog? and where is the cat? This paper seeks to simultaneously exploit the representational capacity of deep networks and the compositional li...

Full description

Saved in:
Bibliographic Details
Published in2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) pp. 39 - 48
Main Authors Andreas, Jacob, Rohrbach, Marcus, Darrell, Trevor, Klein, Dan
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2016
Subjects
Computational modeling
Computer architecture
Dogs
Knowledge discovery
Neural networks
Semantics
Visualization
Online AccessGet full text

Cover

More Information
Summary:Visual question answering is fundamentally compositional in nature-a question like where is the dog? shares substructure with questions like what color is the dog? and where is the cat? This paper seeks to simultaneously exploit the representational capacity of deep networks and the compositional linguistic structure of questions. We describe a procedure for constructing and learning neural module networks, which compose collections of jointly-trained neural "modules" into deep networks for question answering. Our approach decomposes questions into their linguistic substructures, and uses these structures to dynamically instantiate modular networks (with reusable components for recognizing dogs, classifying colors, etc.). The resulting compound networks are jointly trained. We evaluate our approach on two challenging datasets for visual question answering, achieving state-of-the-art results on both the VQA natural image dataset and a new dataset of complex questions about abstract shapes.
ISSN:1063-6919
DOI:10.1109/CVPR.2016.12