M
Mengwei Xu
Researcher at Beijing University of Posts and Telecommunications
Publications - 46
Citations - 982
Mengwei Xu is an academic researcher from Beijing University of Posts and Telecommunications. The author has contributed to research in topics: Android (operating system) & Cloud computing. The author has an hindex of 10, co-authored 46 publications receiving 477 citations. Previous affiliations of Mengwei Xu include Peking University.
Papers
More filters
Proceedings ArticleDOI
DeepCache: Principled Cache for Mobile Deep Vision
TL;DR: DeepCache as mentioned in this paper proposes a principled cache design for deep learning inference in continuous mobile vision, which benefits model execution efficiency by exploiting temporal locality in input video streams and propagates regions of reusable results by exploiting the model's internal structure.
Proceedings ArticleDOI
A First Look at Deep Learning Apps on Smartphones
TL;DR: Zhang et al. as discussed by the authors presented the first empirical study on 16,500 most popular Android apps, demystifying how smartphone apps exploit deep learning in the wild, showing the prosperity of mobile deep learning frameworks and building their cores atop deep learning.
Journal ArticleDOI
DeepWear: Adaptive Local Offloading for On-Wearable Deep Learning
TL;DR: DeepWear as discussed by the authors is a deep learning framework for wearable devices to improve the performance and reduce the energy footprint by offloading DL tasks from a wearable device to its paired handheld device through local network connectivity such as Bluetooth.
Posted Content
A First Look at Deep Learning Apps on Smartphones.
TL;DR: This study presents the first empirical study on 16,500 the most popular Android apps, demystifying how smartphone apps exploit deep learning in the wild and urges optimizations on deep learning models deployed on smartphones, protection of these models, and validation of research ideas on these models.
Journal ArticleDOI
ShuffleDog: Characterizing and Adapting User-Perceived Latency of Android Apps
TL;DR: This paper conducts the first systematic measurement study to quantify the user-perceived latency using typical interaction-intensive Android apps in running with and without background workloads, and designs a new system to address the insufficiency of Android system in ensuring the performance of foreground apps.