Inferring Analogous Attributes     CVPR  2014

Chao-Yeh Chen and Kristen Grauman

Abstract:

The appearance of an attribute can vary considerably from class to class (e.g., a “fluffy” dog vs. a “fluffy” towel), making standard class-independent attribute models break down. Yet, training object-specific models for each attribute can be impractical, and defeats the purpose of using attributes to bridge category boundaries. We propose a novel form of transfer learning that addresses this dilemma. We develop a tensor factorization approach which, given a sparse set of class-specific attribute classifiers, can infer new ones for object-attribute pairs unobserved during training. For example, even though the system has no labeled images of striped dogs, it can use its knowledge of other attributes and objects to tailor “stripedness” to the dog category. With two large-scale datasets, we demonstrate both the need for category-sensitive attributes as well as our method’s successful transfer. Our inferred attribute classifiers perform similarly well to those trained with the luxury of labeled class-specific instances, and much better than those restricted to traditional modes of transfer.

从上图可以看出,通过学习一些特定目标的属性分类器,我们可以类推出相似的属性分类器.该分类器是对目标敏感的,虽然没有特定种类的带标签的训练图像.

1.Introduction:

本文的核心贡献有3点:

1.First, performing transfer jointly in the space of two labeled aspects of the data—namely, categories and attributes—is new. Critically, this means our method is not confined to transfer along same-object or same-attribute boundaries; rather, it discovers analogical relationships based on some mixture of previously seen objects and attributes.

第一点,就是与传统的转移学习不同,本文的转移是联合的转移,即:目标种类和属性的转移.

2.Second, our approach produces a discriminative model for an attribute with zero training examples from that category.

第二点,就是产生一种判别性的模型,尽管该类属性没有训练样本.

3.Third, while prior methods often require information about which classes should transfer to which [2, 29, 26, 1] (e.g., that a motorcycle detector might transfer well to a bicycle), our approach naturally discovers where transfer is possible based on how the observed attribute models relate. It can transfer easily between multiple classes at once, not only pairs, and we avoid the guesswork of manually specifying where transfer is likely.

第三点,就是本文所提出的方法不需要关于什么转移到什么的信息.而可以在多种类别之间很方便的转移.

2. Related Work

In contrast, our approach implicitly discovers analogical relationships among object-sensitive attribute classifiers, and our goal is to generate
novel category-sensitive attribute classifiers.

3. Approach

Given training images labeled by their category and one or more attributes, our method produces as output a series of category-sensitive attribute classifiers. Some of those classifiers are explicitly trained with the labeled data, while the rest are inferred by our method. We show how to create these analogous attribute classifiers via tensor completion.

In the following, we first describe how we train category-sensitive classifiers (Sec. 3.1). Then we define the tensor of attributes (Sec. 3.2) and show how we use it to infer analogous models (Sec. 3.3). Finally, we discuss certain salient aspects of the method design (Sec. 3.4).

3.1. Learning Category-Sensitive Attributes

在现有的系统当中,属性的训练是通过一种种类之间相互独立的方式 ( in a category-independent manner )进行.

在这个工作中,我们挑战传统的训练方式,即:in a completely category-indenpent mannner.

while attributes’ visual cues are often shared among some objects, the sharing is not universal. It can dilute(稀释) the learning process to pool cross-category exemplars indiscriminately. (在某些物体中,属性的视觉线索通常是共享的,但是这种共享不是普遍的.能够非判别性的稀释学习过程来集中跨种类的样本).

一种比较 naive 的做法就是,instead train category-sensitive attributes would be to partition training exemplars by their category labels, and train one attribute per category. 当有足够的 attribute + object combinations 的带标签的样本时,这种策略可能是足够的.但是,初步实验证明该方法是次于训练单个普遍的属性.我们归结了两点原因:

1.even in large-scale collections, the long-tailed distribution of object/scene/attribute occurrences in the real world means that some label pairs will be undersampled, leaving inadequate exemplars to build a statistically sound model,

2.this naive approach completely ignores attributes’ inter-class semantic ties. 属性类别之间的语意连接.

To overcome these shortcomings, we instead use an importance-weighted support vector machine (SVM) to train each category-sensitive attribute. 每一个训练样本(xi, yi)都包括一个图像描述xi,和标签yi 属于{-1, +1}.

论文阅读之 Inferring Analogous Attributes CVPR 2014的更多相关文章

  1. 目标检测--Rich feature hierarchies for accurate object detection and semantic segmentation(CVPR 2014)

    Rich feature hierarchies for accurate object detection and semantic segmentation 作者: Ross Girshick J ...

  2. 目标检测--Scalable Object Detection using Deep Neural Networks(CVPR 2014)

    Scalable Object Detection using Deep Neural Networks 作者: Dumitru Erhan, Christian Szegedy, Alexander ...

  3. 从CVPR 2014看计算机视觉领域的最新热点

    编者按:2014年度计算机视觉方向的顶级会议CVPR上月落下帷幕.在这次大会中,微软亚洲研究院共有15篇论文入选.今年的CVPR上有哪些让人眼前一亮的研究,又反映出哪些趋势?来听赴美参加会议的微软亚洲 ...

  4. 从CVPR 2014看计算机视觉领域的最新热点

    2014看计算机视觉领域的最新热点" title="从CVPR 2014看计算机视觉领域的最新热点"> 编者按:2014年度计算机视觉方向的顶级会议CVPR上月落下 ...

  5. 论文阅读(Xiang Bai——【arXiv2016】Scene Text Detection via Holistic, Multi-Channel Prediction)

    Xiang Bai--[arXiv2016]Scene Text Detection via Holistic, Multi-Channel Prediction 目录 作者和相关链接 方法概括 创新 ...

  6. DeconvNet 论文阅读理解

    学习语义分割反卷积网络DeconvNet 一点想法:反卷积网络就是基于FCN改进了上采样层,用到了反池化和反卷积操作,参数量2亿多,非常大,segnet把两个全连接层去掉,效果也能很好,显著减少了参数 ...

  7. 论文阅读笔记四十四:RetinaNet:Focal Loss for Dense Object Detection(ICCV2017)

    论文原址:https://arxiv.org/abs/1708.02002 github代码:https://github.com/fizyr/keras-retinanet 摘要 目前,具有较高准确 ...

  8. 论文阅读:Review of Visual Saliency Detection with Comprehensive Information

    这篇文章目前发表在arxiv,日期:20180309. 这是一篇针对多种综合性信息的视觉显著性检测的综述文章. 注:有些名词直接贴原文,是因为不翻译更容易理解.也不会逐字逐句都翻译,重要的肯定不会错过 ...

  9. 论文阅读笔记三十七:Grid R-CNN(CVPR2018)

    论文源址:https://arxiv.org/abs/1811.12030 开源代码:未公开 摘要 本文提出了目标检测网络Grid R-CNN,其基于网格定位机制实现准确的目标检测.传统方法主要基于回 ...

随机推荐

  1. iOS中属性Property的常用关键字的使用说明

    属性关键字的作用 现在我们iOS开发中,基本都是使用ARC(自动引用计数)技术,来编写我们的代码.因此在属性property中我们经常使用的关键字有strong,weak,assign,copy,no ...

  2. Smart Card Filesystem

  3. mysql 批量创建表,利用存储过程

    最近根据需求,需要提前创建一批日志表,以日期结尾,每天创建一张,例如XXX20160530,请参考如下: BEGIN    DECLARE `sName` VARCHAR(128);   DECLAR ...

  4. 修复win8引导

    格式化那个350MB的分区(Win8安装盘启动之后挂载在c:之后,用Win8的安装U盘,进去修复模式,然后进入高级选项的命令行提示符模式.接着,转到安装现有Win8的分区(Win8安装盘启动之后挂载在 ...

  5. 51nod 最长公共子序列Lcs

    有深入 了解了一点 .  51nod 可以用来加深 算法理解程度 ,

  6. dfs介绍

    深度优先搜索(DFS) [算法入门] 郭志伟@SYSU:raphealguo(at)qq.com 2012/05/12 1.前言 深度优先搜索(缩写DFS)有点类似广度优先搜索,也是对一个连通图进行遍 ...

  7. Could not launch "app_name"

    真机测试 不报错 编译通过后 Xcode总出这个错 process launch faild:NotFound-------解决办法 :重启设备

  8. 浅谈__declspec(dllexport)和__declspec(dllimport)

    __declspec(dllimport)和__declspec(dllexport)经常是成对的,在动态链接库中__declspec(dllexport)导出dll中的成员,__declspec(d ...

  9. Ubuntu安装Sun JDK及如何设置默认java JDK

    假设系统为Ubuntu 10.10 64位.    先来看看当前的系统默认的java版本,及提供商.通过命令java -version       kemix@kemix-VB10-64:~$ jav ...

  10. Day04_JAVA语言基础第四天

    1.循环(掌握) 1.什么时候使用(理解) 如果我们发现有很多重复内容的时候就要使用循环 2.好处(理解) 让我们的代码看起来更精炼了 3.循环的组成(理解) 1 初始化条件:一般定义的是一个初始变量 ...