@inproceedings{4fa609e695bc4948ba52d9d4134f0019,
title = "Optimization of Remote Desktop with CNN-based Image Compression Model",
abstract = "Remote desktop systems become commonly used for users to enhance the efficiency of their daily tasks commonly. In this work, we propose an expanded image compression model with convolutional neural network (CNN) and train two jointly optimized CNN based models as the image encoder and decoder to optimize the compression of the desktop images and design a new compartmentalization of the update desktop region to fit the CNN encoder. We implement the proposed encoding on the open source Remote Frame Buffer (RFB) protocol. Compared with tight encoding which is dedicated to low-bandwidth remote desktop, the proposed encoding method prompts the user experience with a even lower network bandwidth consumption.",
keywords = "Convolutional neural network, Remote desktop, Remote frame buffer, VNC",
author = "Hejun Wang and Hongjun Dai and Meikang Qiu and Meiqin Liu",
note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 14th International Conference on Knowledge Science, Engineering and Management, KSEM 2021 ; Conference date: 14-08-2021 Through 16-08-2021",
year = "2021",
doi = "10.1007/978-3-030-82136-4\_56",
language = "英语",
isbn = "9783030821357",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Science and Business Media Deutschland GmbH",
pages = "692--703",
editor = "Han Qiu and Cheng Zhang and Zongming Fei and Meikang Qiu and Sun-Yuan Kung",
booktitle = "Knowledge Science, Engineering and Management - 14th International Conference, KSEM 2021, Proceedings",
}