READING NOTE: PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection

原创 2016年08月30日 20:58:12

TITLE: PVANET Deep but Lightweight Neural Networks for Real-time Object Detection

AUTHER: Kye-Hyeon Kim, Yeongjae Cheon, Sanghoon Hong, Byungseok Roh, Minje Park

ASSOCIATION: Intel Imaging and Camera Technology

FROM: arXiv:1608.08021

CONTRIBUTIONS

  1. An efficient object detector based on CNN is proposed, which has the following advantages:
    • Computational cost: 7.9GMAC for feature extraction with 1065x640 input (cf. ResNet-101: 80.5GMAC1)
    • Runtime performance: 750ms/image (1.3FPS) on Intel i7-6700K CPU with a single core; 46ms/image (21.7FPS) on NVIDIA Titan X GPU
    • Accuracy: 81.8% mAP on VOC-2007; 82.5% mAP on VOC-2012 (2nd place)

Method

The author utilizes the pipline of Faster-RCNN, which is “CNN feature extraction + region proposal + RoI classification”. The author claims that feature extraction part needs to be redesigned, since region proposal part is not computationally expensive and classification part can be efficiently compressed with common techniques like truncated SVD. And the principle is “less channels with more layers” and adoption of some building blocks including concatenated ReLU, Inception, and HyperNet. The structure of the network is as follows:

Some Details

  1. Concatenated rectified linear unit (C.ReLU) is applied to the early stage of the CNNs (i.e., first several layers from the network input) to reduce the number of computations by half without losing accuracy. In my understanding, the C.ReLU encourages the network to learn Gabor-like filters and helps to accelerate the forward-propagation. If the output of the C.ReLu is 64, its convolution layer only needs 32-channel outputs. And it may harm the performance if it is used to the later stage of the CNNs, because it keeps the negative responses as activated signal, which means that a mad brain is trained.
  2. Inception is applied to the remaining of the feature generation sub-network. An Inception module produces output activations of different sizes of receptive fields, so that increases the variety of receptive field sizes in the previous layer. All the design policies can be found in this related work.
  3. The author adopted the idea of multi-scale representation like HyperNet that combines several intermediate outputs so that multiple levels of details and non-linearities can be considered simultaneously. Direct concatenation of all abstraction layers may produce redundant information with much higher compute requirement and layers which are too early for object proposal and classification would be little help. The author combines 1) the last layer and 2) two intermediate layers whose scales are 2x and 4x of the last layer, respectively.
  4. Residual structure is also used in this network, which helps to train very deep CNNs.
版权声明:本文为博主原创文章,未经博主允许不得转载。欢迎访问博主个人主页 http://joshua881228.webfactional.com/

ubuntu修改gedit的tab键为4个空格

1、要修改gedit的tab键需要在菜单项中修改,如果是在Terminal中打开的gedit是不显示菜单的,所以需要右键点击一个文件,通过gedit打开,这是能够看到菜单栏; 2、通常ubuntu应...

ubuntu16 caffe GPU

第一部分,准备材料(NVIDIA官网下载):显卡驱动àNVIDIA-Linux-x86_64-367.44.runCuda8.0àcuda_8.0.27_linux.run网址:https://dev...

PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection - arxiv 2016.08

PVANET: Deep but Lightweight Neural Networks forReal-time Object Detection, arxiv 16.08 论文地址:https:/...

论文笔记:PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection

引入PVANET是韩国研究者(Intel Imaging and Camera Technology)结合了最新的技术,在TITAN X上实现了轻量级模型的实施目标检测任务,在PASCAL VOC上取...

PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection

PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection论文地址 github源码Introductio...

Deep Neural Networks for Object Detection论文翻译

这是2013年使用深度学习做目标检测的一篇论文,相对于经典的rcnn是要出现的更早一些,这篇博客将对这篇论文进行一些翻译和解读,若理解有偏差和错误还望指出,大家互相交流和学习。 论文题目:Deep ...
  • xbcReal
  • xbcReal
  • 2017年07月15日 00:45
  • 515

Deep Neural Networks for Object Detection(基于DNN的对象检测)

本博客对论文《Deep Neural Networks for object detection》进行了翻译

论文笔记:Faster R-CNN:Towards Real-Time Object Detection with Region Proposal Networks

转自:http://blog.csdn.net/anxiaoxi45/article/details/46522155 论文笔记:Faster R-CNN:Towards Real-Time ...
  • csyhhb
  • csyhhb
  • 2016年01月05日 10:37
  • 858

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks(更快的RCNN:通过区域提议网络实现实时

Faster R-CNN: Towards Real-Time Object Detection with Region ProposalNetworks Shaoqing Ren, Kai...

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks----论文笔记

一、为什么提出Faster R-CNNFaster R-CNN的前身Fast R-CNN能达到实时检测,如果不考虑它用selective search计算region proposal所花时间的话。为...
内容举报
返回顶部
收藏助手
不良信息举报
您举报文章:READING NOTE: PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection
举报原因:
原因补充:

(最多只允许输入30个字)