READING NOTE: Rethinking the Inception Architecture for Computer Vision
来源:互联网 发布:淘宝直通车没流量 编辑:程序博客网 时间:2024/05/16 07:13
TITLE: Rethinking the Inception Architecture for Computer Vision
AUTHER: Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jonathon Shlens, Zbigniew Wojna
ASSOCIATION: Google Inc., University College London
FROM: arXiv:1512.00567
CONTRIBUTIONS
- Several general and specific design priciples are discussed
Design Choices
General Design Principles
- Avoid representational bottlenecks, especially early in the network. One should avoid bottlenecks with extreme compression. In general the representation size should gently decrease from the inputs to the outputs before reaching the final representation used for the task at hand.
- Higher dimensional representations are easier to process locally within a network. Increasing the activations per tile in a convolutional network allows for more disentangled features. The resulting networks will train faster.
- Spatial aggregation can be done over lower dimensional embeddings without much or any loss in representational power.
- Balance the width and depth of the network.
Sepecific Design Strategy
- Factorizing Convolutions with Large Filter Size includes Factorization into smaller convolutions and Spatial Factorization into Asymmetric Convolutions. Both help to improve the speed and the complexity of the learnt function.
- Utility of Auxiliary Classifiers act as regularizer rather than help evolving the low-level features. Near the end of training, the network with the auxiliary branches starts to overtake the accuracy of the network without any auxiliary branch and reaches a slightly higher plateau.
- Efficient Grid Size Reduction reduces the computational cost while removing the representational bottleneck.
Some Other Ideas
In this paper, a very intereting experiment is of value to be noted. With different perceptive field size, the networks can achieve similar results if similar computational cost is constant.
In my own trials of SSD, I found networks of similar computational cost with differnt perceptive field size have very different result in detection task. For example, Network A has a perceptive field size of 112x112, while Network B is 170x170. Network B has a slightly better performance on classificatino task on Network A. On the contrary, after the two networks are finetuned on 200*200 images on detection task, Network A is better. Thus, how about we train a network with the perceptive field size of, let’s say, 56x56 and finetune it on 100x100 images? Will it have a comparable result?
- READING NOTE: Rethinking the Inception Architecture for Computer Vision
- rethinking the inception architecture for computer vision
- Rethinking the Inception Architecture for Computer Vision
- Inception-v3:"Rethinking the Inception Architecture for Computer Vision"
- Inception V3(Rethinking the Inception Architecture for Computer Vision)
- Rethinking the Inception Architecture for Computer Vision 笔记
- 论文笔记 | Rethinking the Inception Architecture for Computer Vision
- 《Rethinking the Inception Architecture for Computer Vision》笔记
- 网络模型 Inception V2/V3-Rethinking the Inception Architecture for Computer Vision
- googLeNet--Inception四部曲三Rethinking the Inception Architecture for Computer Vision
- [深度学习论文笔记][Image Classification] Rethinking the Inception Architecture for Computer Vision
- Inception系列2_Rethinking the Inception Architecture for Computer Vision
- Reading List for Computer Vision Newbie
- Reading Lists for Advanced Computer Vision in 2009 and 2010
- Reading Note : Gibbs Sampling for the Uninitiated
- Reading Note : Gibbs Sampling for the Uninitiated
- Reading Note: MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- Flexible Models for Computer Vision
- java中String 的==和equals;String直接赋值和使用new String()
- Java常考面试题
- 坚持#第24天~回黄陂了
- Intent学习(1)
- How To Use Codelite
- READING NOTE: Rethinking the Inception Architecture for Computer Vision
- python(五)常用模块学习
- Android studio中控制台输出乱码
- 面向对象知识点回顾
- Leetcode 136. Single Number
- 18. 4Sum
- Section 1.3 Greedy Algorithm
- AJAX实现异步提交
- idea 打包包含依赖的JAR文件