Show and Tell: Lessons learned from the 2015 MSCOCO Image Captioning Challenge

Automatically describing the content of an image is a fundamental problem in artificial intelligence that connects computer vision and natural language processing. In this paper, we present a generative model based on a deep recurrent architecture that combines recent advances in computer vision and...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2016-09
Main Authors: Vinyals, Oriol, Toshev, Alexander, Bengio, Samy, Dumitru Erhan
Format: Article
Language:English
Subjects:
Artificial intelligence
Competition
Computer vision
Machine translation
Model accuracy
Natural language processing
Sentences
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Automatically describing the content of an image is a fundamental problem in artificial intelligence that connects computer vision and natural language processing. In this paper, we present a generative model based on a deep recurrent architecture that combines recent advances in computer vision and machine translation and that can be used to generate natural sentences describing an image. The model is trained to maximize the likelihood of the target description sentence given the training image. Experiments on several datasets show the accuracy of the model and the fluency of the language it learns solely from image descriptions. Our model is often quite accurate, which we verify both qualitatively and quantitatively. Finally, given the recent surge of interest in this task, a competition was organized in 2015 using the newly released COCO dataset. We describe and analyze the various improvements we applied to our own baseline and show the resulting performance in the competition, which we won ex-aequo with a team from Microsoft Research, and provide an open source implementation in TensorFlow.
ISSN:2331-8422
DOI:10.48550/arxiv.1609.06647