Microsoft COCO Captions: Data Collection and Evaluation Server

Microsoft COCO Captions: Data Collection and Evaluation Server
复制标题

DOI:
--
复制
发表时间:
2015-04
期刊:
ArXiv
影响因子:
--
通讯作者:
Xinlei Chen;Hao Fang;Tsung-Yi Lin;Ramakrishna Vedantam;Saurabh Gupta;Piotr Dollár;C. L. Zitnick
Xinlei Chen;Hao Fang;Tsung-Yi Lin;Ramakrishna Vedantam;Saurabh Gupta;Piotr Dollár;C. L. Zitnick
中科院分区:
其他
文献类型:
--
作者:
Xinlei Chen;Hao Fang;Tsung-Yi Lin;Ramakrishna Vedantam;Saurabh Gupta;Piotr Dollár;C. L. Zitnick

文献摘要

被引文献

相似文献

在本文中,我们描述了Microsoft COCO Caption数据集和评估服务器。完成后,数据集将包含超过150万个描述超过330,000张图像的标题。对于训练和验证图像,将提供五个独立的人工生成的标题。为了确保自动字幕生成算法评估的一致性,使用评估服务器。评估服务器接收候选字幕,并使用几个流行的指标(包括BLEU、METEOR、ROUGE和CIDER)对其进行评分。提供了使用评估服务器的说明。
In this paper we describe the Microsoft COCO Caption dataset and evaluation server. When completed, the dataset will contain over one and a half million captions describing over 330,000 images. For the training and validation images, five independent human generated captions will be provided. To ensure consistency in evaluation of automatic caption generation algorithms, an evaluation server is used. The evaluation server receives candidate captions and scores them using several popular metrics, including BLEU, METEOR, ROUGE and CIDEr. Instructions for using the evaluation server are provided.