Corpus ID: 6621072

Are Very Deep Neural Networks Feasible on Mobile Devices

  title={Are Very Deep Neural Networks Feasible on Mobile Devices},
  author={S. Rallapalli and Hang Qiu and A. J. Bency and S. Karthikeyan and R. Govindan and B. S. Manjunath and R. Urgaonkar},
  • S. Rallapalli, Hang Qiu, +4 authors R. Urgaonkar
  • Published 2016
  • Computer Science
  • In the recent years, the computing power of mobile devices has increased tremendously, a trend that is expected to continue in the future. [...] Key Method We then quantify the performance of several deep CNNspecific memory management techniques, some of which leverage the observation that these CNNs have a small number of layers that require most of the memory. We find that a particularly novel approach that offloads these bottleneck layers to the mobile device’s CPU and pipelines frame processing is a…Expand Abstract
    21 Citations
    Exploring the Capabilities of Mobile Devices Supporting Deep Learning
    • 6
    CrowdVision: A Computing Platform for Video Crowdprocessing Using Deep Learning
    • 3
    • PDF
    Deep Neural Mobile Networking
    • Highly Influenced
    • PDF
    Deep Learning in Mobile and Wireless Networking: A Survey
    • 414
    • Highly Influenced
    • PDF
    Deep learning hashing for mobile visual search
    • 24
    Optimizing Deep Learning Inference on Embedded Systems Through Adaptive Model Selection
    • 14
    • PDF


    Can Deep Learning Revolutionize Mobile Sensing?
    • 188
    • PDF
    Caffe: Convolutional Architecture for Fast Feature Embedding
    • 12,471
    • PDF
    Very Deep Convolutional Networks for Large-Scale Image Recognition
    • 43,505
    • PDF
    ImageNet classification with deep convolutional neural networks
    • 58,143
    • PDF
    VAST: The illusion of a large memory space for GPUs
    • 32
    • PDF
    Balancing energy, latency and accuracy for mobile sensor data classification
    • 110
    • PDF
    GPUswap: Enabling Oversubscription of GPU Memory through Transparent Swapping
    • 30
    ORBIT: a smartphone-based platform for data-intensive embedded sensing applications
    • 12
    • PDF
    Fast R-CNN
    • Ross B. Girshick
    • Computer Science
    • 2015 IEEE International Conference on Computer Vision (ICCV)
    • 2015
    • 9,835
    • PDF