# Optimizing Neural Network for Computer Vision task in Edge Device

arXiv: https://arxiv.org/abs/2110.00791 · HTML: https://mlsanity.com/paper/2110.00791/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 28 references that are in the map.
- First posted on arXiv 2 Oct 2021.

## Builds on (most cited first, 11 of 11)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems](https://mlsanity.com/paper/1603.04467/) (2016)
- [Super-Convergence: Very Fast Training of Neural Networks Using Large Learning Rates](https://mlsanity.com/paper/1708.07120/) (2017)
- [All you need is a good init](https://mlsanity.com/paper/1511.06422/) (2015)
- [Large-Batch Training for LSTM and Beyond](https://mlsanity.com/paper/1901.08256/) (2019)

## Abstract (arXiv)

The field of computer vision has grown very rapidly in the past few years due to networks like convolution neural networks and their variants. The memory required to store the model and computational expense are very high for such a network limiting it to deploy on the edge device. Many times, applications rely on the cloud but that makes it hard for working in real-time due to round-trip delays. We overcome these problems by deploying the neural network on the edge device itself. The computational expense for edge devices is reduced by reducing the floating-point precision of the parameters in the model. After this the memory required for the model decreases and the speed of the computation increases where the performance of the model is least affected. This makes an edge device to predict from the neural network all by itself.