Recognizing the Order of Four-Scene Comics by Evolutionary Deep Learning
Recognizing the Order of Four-Scene Comics by Evolutionary Deep Learning
复制标题
DOI：
10.1007/978-3-319-94649-8_17
复制
发表时间：
2018-06
期刊：
影响因子：
--
通讯作者：
Saya Fujino;N. Mori;Keinosuke Matsumoto
中科院分区：
文献类型：
--
作者：
Saya Fujino;N. Mori;Keinosuke Matsumoto
近年来，漫画分析已成为人工智能领域一个颇具吸引力的研究课题。在这项研究中，我们专注于四场景漫画，并将深度卷积神经网络（DCNN）应用于数据以理解顺序结构。 DCNN 超参数的调整需要付出相当大的努力。为了解决这个问题，我们通过遗传算法提出了一种称为进化深度学习（evoDL）的新方法。 evoDL 的有效性已通过一项旨在识别实际四场景漫画中的结构问题的实验得到证实。
In recent years, comic analysis has become an attractive research topic in the field of artificial intelligence. In this study, we focused on the four-scene comics and applied deep convolutional neural networks (DCNNs) to the data for understanding the order structure. The tuning of the DCNN hyperparameters requires considerable effort. To solve this problem, we propose a novel method called evolutionary deep learning (evoDL) by means of genetic algorithms. The effectiveness of evoDL is confirmed by an experiment conducted to identify structural problems in actual four-scene comics.