Host: The Japan Society of Mechanical Engineers
Name : [in Japanese]
Date : June 05, 2019 - June 08, 2019
In this paper, the authors aim at embedding and restoring arbitrary information in an arbitrary image using CNN. In order to achieve this objective, we propose a model composed of two CNNs of different roles of an Encoder CNN that embeds information on an image and a Decoder CNN that restores information embedded in the image. In the proposed method, the QR code is used as the medium of information to be embedded. Thanks to the error correction nature of QR code, embedded information is expected to be restored without errors.
