1. pdf

ParallelFusion: Towards Maximum Utilization of Mobile GPU for DNN Inference

1.1. Association for Computing Machinery

ParallelFusion: Towards Maximum Utilization of Mobile GPU for DNN Inference
https://dl.acm.org/doi/abs/10.1145/3469116.3470014

1.2. 清华大学智能产业研究院 (Institute for AI Industry Research, Tsinghua University，AIR, THU)

ParallelFusion: Towards Maximum Utilization of Mobile GPU for DNN Inference
https://air.tsinghua.edu.cn/pdf/ParallelFusion-Towards-Maximum-Utilization-of-Mobile-GPU-for-DNN-Inference.pdf

1.3. Publications

https://jingyulee.com/

ParallelFusion: Towards Maximum Utilization of Mobile GPU for DNN Inference

2. ACM MobiSys 2021

https://www.sigmobile.org/mobisys/2021/

The 19th ACM International Conference on Mobile Systems, Applications, and Services.

ACM MobiSys 2021 seeks to present innovative and significant research on the design, implementation, usage, and evaluation of mobile computing and wireless systems, applications, and services. This conference builds on the success of the previous seventeen ACM MobiSys conferences. It is sponsored by ACM SIGMOBILE.

2.1. Technical Program - 25 June 2021

https://emdl21.github.io/program.html

Session 1 - Efficient Mobile DNN Inference

2.2. Full presentation

ParallelFusion: Towards Maximum Utilization of Mobile GPU for DNN Inference
Jingyu Lee (Seoul National University), Yunxin Liu (Institute for AI Industry Research (AIR), Tsinghua University), Youngki Lee (Seoul National University)
https://www.youtube.com/watch?v=vndficgb16s

References

[1] Yongqiang Cheng, https://yongqiang.blog.csdn.net/