Perceptual Generative Adversarial Networks for Small Object Detection

2017-07-11  19:47:46   CVPR 2017

  This paper use GAN to handle the issue of small object detection which is a very hard problem in general object detection. As shown in the following figures, small object and large objects usually shown different representations from the feature level.   

  Thus, it is possbile to use Percetual GAN to super-resolution of feature maps of small objects to obtain better detection performance.

  It consists of two subnetworks, i.e., a generator network and a perceptual discriminator network. Specifically, the generator is a deep residual based feature generative model which transforms the original poor features of small objects to highly discriminative ones by introducing fine-grained details from lower-level layers, achieving “super-resolution” on the intermediate representations.

  

  Different from normal GAN, this network also introduce a new perceptual loss tailored from the detection purpose. That is to say, the discriminator not only need to deal with the adversarial loss, but also need to justify the detection accuray benefiting from the generated super-resolved features with a perceptual loss.

  The proposed contributions:

  (1) We are the first to successfully apply GAN-alike models to solve the challenging small-scale object detection problems.

   (2) We introduce a new conditional generator model that learns the additive residual representation between large and small objects, instead of generating the complete representations as before.

   (3) We introduce a new perceptual discriminator that provides more comprehensive supervision beneficial for detections, instead of barely differentiating fake and real.

   (4) Successful applications on traffic sign detection and pedestrian detection have been achieved with the state-of-the-art performance.

  Figure 2. Training procedure of object detection network based on the Perceptual GAN.

  

  As shown in Figure 2, the generator network aims to generate super-resoved representation for the small object.

  The discriminator includes two branches, i.e.

    1. the adversarial branch  for differentiating between the generated superresolved representation.

    2. the perception branch for justifying the detection accurcy benefiting from the generation representation. 


  ==>> Dicriminative Network Architecture:

  The D network need to justify the dection accuracy benefiting from the generated super-resovled feature.

  Given the adversarial loss $L_{dis_a}$ and the perceptual loss $L_{dis_p}$ , a final loss function Ldis can be produced as weighted sum of both individual loss components. Given weighting parameters w1 and w2, we define Ldis = w1 × Ldis_a + w2 × Ldis_p to encourage the generator network to generate super-resolved representation with high detection accuracy. Here we set both w1 and w2 to be one. 

    


  


  

  

Paper Reading: Perceptual Generative Adversarial Networks for Small Object Detection的更多相关文章

  1. 【文献阅读】Perceptual Generative Adversarial Networks for Small Object Detection –CVPR-2017

    Perceptual Generative Adversarial Networks for Small Object Detection 2017CVPR 新鲜出炉的paper,这是针对small ...

  2. Perceptual Generative Adversarial Networks for Small Object Detection

    Perceptual Generative Adversarial Networks for Small Object Detection 感知生成对抗网络用于目标检测 论文链接:https://ar ...

  3. (转)Introductory guide to Generative Adversarial Networks (GANs) and their promise!

    Introductory guide to Generative Adversarial Networks (GANs) and their promise! Introduction Neural ...

  4. 论文笔记之:UNSUPERVISED REPRESENTATION LEARNING WITH DEEP CONVOLUTIONAL GENERATIVE ADVERSARIAL NETWORKS

    UNSUPERVISED REPRESENTATION LEARNING WITH DEEP CONVOLUTIONAL GENERATIVE ADVERSARIAL NETWORKS  ICLR 2 ...

  5. Generative Adversarial Networks overview(2)

    Libo1575899134@outlook.com Libo (原创文章,转发请注明作者) 本文章会先从Gan的简单应用示例讲起,从三个方面问题以及解决思路覆盖25篇GAN论文,第二个大部分会进一步 ...

  6. Generative Adversarial Networks overview(1)

    Libo1575899134@outlook.com Libo (原创文章,转发请注明作者) 本文章会先从Gan的简单应用示例讲起,从三个方面问题以及解决思路覆盖25篇GAN论文,第二个大部分会进一步 ...

  7. GD-GAN: Generative Adversarial Networks for Trajectory Prediction and Group Detection in Crowds

    GD-GAN: Generative Adversarial Networks for Trajectory Prediction and Group Detection in Crowds 2019 ...

  8. Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

    Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks 2019-06-01 09:52:4 ...

  9. 文献阅读报告 - Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networks

    paper:Gupta A , Johnson J , Fei-Fei L , et al. Social GAN: Socially Acceptable Trajectories with Gen ...

随机推荐

  1. Cipher

    Description Bob and Alice started to use a brand-new encoding scheme. Surprisingly it is not a Publi ...

  2. shell 编程每日100行

    xiluhua@localhost ~/tscripts $ echo "hello world !" -bash: !": event not found xiluhu ...

  3. Qt浅谈之一:内存泄露(总结)

    一.简介       Qt内存管理机制:Qt 在内部能够维护对象的层次结构.对于可视元素,这种层次结构就是子组件与父组件的关系:对于非可视元素,则是一个对象与另一个对象的从属关系.在 Qt 中,在 Q ...

  4. Linux 下wifi 驱动开发(四)—— USB接口WiFi驱动浅析

    源: Linux 下wifi 驱动开发(四)—— USB接口WiFi驱动浅析

  5. JavaScript快速总结之一

    js格式和位置: <!--js内容可以在head中,也可以在body中,不同位置执行顺序会有区别 ,另外也可以单独放到一个js文件中,这样方便维护和修改,而且,可以加快html页面的加载速度.- ...

  6. selenium webdriver 实现Canvas画布自动化测试

    https://blog.csdn.net/xiaoguanyusb/article/details/80324210 由借鉴意义, 转过来 canvas 是一个画布,定位元素时只能定位到画布上,如下 ...

  7. Fiddler(三)Fiddler 报错creation of the root certificate was not successful

    打开CMD,找到Fiddler所在目录,我这里是把汉化版的解压在了桌面,所以通过CMD进入桌面下的Fiddler文件夹. 执行命令 makecert.exe -r -ss my -n -h -cy a ...

  8. keepalived与zookeeper

    keepalived与zookeeper都可以用来实现高可用,高可用一般跟负载均衡会一起考虑,所以通常也会考虑到相应的负载均衡能力, 1.概括对比: 1.1.Keepalived: 优点:简单,基本不 ...

  9. Mysql 存储过程中使用多游标

    Mysql 存储过程中使用多游标 drop procedure IF EXISTS test_proc_1; create procedure test_proc_1() begin ; ) ; ) ...

  10. plsql developer 10注册码----亲测截止2017年5月6可用

    亲测截止2017年5月6可用 Product Code:4t46t6vydkvsxekkvf3fjnpzy5wbuhphqzserial Number:601769password:xs374ca