Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/162632
Title: Learning to compose and reason with language tree structures for visual grounding
Authors: Hong, Richang
Liu, Daqing
Mo, Xiaoyu
He, Xiangnan
Zhang, Hanwang
Keywords: Engineering::Electrical and electronic engineering
Engineering::Computer science and engineering
Issue Date: 2019
Source: Hong, R., Liu, D., Mo, X., He, X. & Zhang, H. (2019). Learning to compose and reason with language tree structures for visual grounding. IEEE Transactions On Pattern Analysis and Machine Intelligence, 44(2), 684-696. https://dx.doi.org/10.1109/TPAMI.2019.2911066
Journal: IEEE Transactions on Pattern Analysis and Machine Intelligence
Abstract: Grounding natural language in images, such as localizing "the black dog on the left of the tree", is one of the core problems in artificial intelligence, as it needs to comprehend the fine-grained language compositions. However, existing solutions merely rely on the association between the holistic language features and visual features, while neglect the nature of composite reasoning implied in the language. In this paper, we propose a natural language grounding model that can automatically compose a binary tree structure for parsing the language and then perform visual reasoning along the tree in a bottom-up fashion. We call our model RvG-Tree: Recursive Grounding Tree, which is inspired by the intuition that any language expression can be recursively decomposed into two constituent parts, and the grounding confidence score can be recursively accumulated by calculating their grounding scores returned by the two sub-trees.RvG-Tree can be trained end-to-end by using the Straight-Through Gumbel-Softmax estimator that allows the gradients from the continuous score functions passing through the discrete tree construction. Experiments on several benchmarks show that our model achieves the state-of-the-art performance with more explainable reasoning.
URI: https://hdl.handle.net/10356/162632
ISSN: 0162-8828
DOI: 10.1109/TPAMI.2019.2911066
Rights: © 2019 IEEE. All rights reserved.
Fulltext Permission: none
Fulltext Availability: No Fulltext
Appears in Collections:EEE Journal Articles
SCSE Journal Articles

SCOPUSTM   
Citations 20

8
Updated on Nov 25, 2022

Web of ScienceTM
Citations 20

8
Updated on Nov 24, 2022

Page view(s)

9
Updated on Nov 25, 2022

Google ScholarTM

Check

Altmetric


Plumx

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.