Do Larger (More Accurate) Deep Neural Network Models Help in Edge-assisted Augmented Reality?
Do Larger (More Accurate) Deep Neural Network Models Help in Edge-assisted Augmented Reality?
复制标题
更大（更准确）的深度神经网络模型有助于边缘辅助增强现实吗？
DOI：
10.1145/3472727.3472807
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Hu, Y. Charlie
中科院分区：
文献类型：
--
作者：
Meng, Jiayi;Kong, Zhaoning;Xu, Qiang;Hu, Y. Charlie
边缘辅助增强现实（AR）将基于计算密集型深度神经网络（DNN）的AR任务卸载到边缘服务器，面临着一个重要的设计挑战：如何从为每个AR任务提出的许多选择中选择DNN模型进行卸载。对于每个AR任务，例如，深度估计，随着时间的推移，已经提出了许多基于DNN的模型，这些模型在准确性和复杂性方面有所不同。一般来说，更精确的模型也更复杂;它们更大，推理时间更长。因此，在卸载中选择较大的模型可以为卸载的帧提供更高的精度，但也会导致更长的周转时间，在此期间，AR应用程序必须重用最后卸载的帧的估计结果，这可能导致平均精度较低。我们设计了最佳的卸载时间表，并进一步考虑了众多因素的影响，如设备上的快速跟踪，帧缩小和可用的网络带宽。我们的研究结果表明，对于边缘辅助的单目深度估计，通过适当的帧缩小和快速跟踪，与小模型相比，大模型的准确性提高可以抵消其较长的周转时间，从而在LTE和5G毫米波下提供更高的跨帧平均估计准确性。
Edge-assisted Augmented Reality (AR) which offloads compute-intensive Deep Neural Network (DNN)-based AR tasks to edge servers faces an important design challenge: how to pick the DNN model out of many choices proposed for each AR task for offloading. For each AR task, e.g., depth estimation, many DNN-based models have been proposed over time that vary in accuracy and complexity. In general, more accurate models are also more complex; they are larger and have longer inference time. Thus choosing a larger model in offloading can provide higher accuracy for the offloaded frames but also incur longer turnaround time, during which the AR app has to reuse the estimation result from the last offloaded frame, which can lead to lower average accuracy.In this paper, we experimentally study this design tradeoff using depth estimation as a case study. We design optimal offloading schedule and further consider the impact of numerous factors such as on-device fast tracking, frame downsizing and available network bandwidth. Our results show that for edge-assisted monocular depth estimation, with proper frame downsizing and fast tracking, compared to small models, the improved accuracy of large models can offset its longer turnaround time to provide higher average estimation accuracy across frames under both LTE and 5G mmWave.