[CVPR 2016] Weakly Supervised Deep Detection Networks论文笔记
Weakly Supervised Deep Detection Networks,Hakan Bilen,Andrea Vedaldi
https://www.cv-foundation.org/openaccess/content_cvpr_2016/papers/Bilen_Weakly_Supervised_Deep_CVPR_2016_paper.pdf
亮点
- 把弱监督检测问题解释为proposal排序的问题,通过比较所有proposal的类别分数得到一个比较正确的排序,这种思想与检测中评测标准的计算方法一致
相关工作
The MIL strategy results in a non-convex optimization problem; in practice, solvers tend to get stuck in local optima
such that the quality of the solution strongly depends on the initialization.
- developing various initialization strategies [19, 5, 32, 4]
- [19] propose a self-paced learning strategy
- [5] initialize object locations based on the objectness score.
- [4] propose a multi-fold split of the training data to escape local optima.
- on regularizing the optimization problem [31, 1].
- [31] apply Nesterov’s smoothing technique to the latent SVM formulation
- [1] propose a smoothed version of MIL that softly labels object instances instead of choosing the highest scoring ones.
- Another line of research in WSD is based on the idea of identifying the similarity between image parts.
- [31] propose a discriminative graph-based algorithm that selects a subset of windows such that each window is connected to its nearest neighbors in positive images.
- [32] extend this method to discover multiple co-occurring part configurations.
- [36] propose an iterative technique that applies a latent semantic clustering via latent Semantic Analysis (pLSA)
- [2] propose a formulation that jointly learns a discriminative model and enforces the similarity of the selected object regions via a discriminative convex clustering algorithm
方法
本文采用的方法非常简单易懂,主要分为以下三部:
- 将特征和region proposal的结果输入spatial pyramid pooling层,取出与区域相关的特征向量,并输入两个fc层
- 分类:fc层的输出通过softmax分类器,计算出这一区域类别
- 检测:fc层的输出通过softmax分类器,与上面不同的是归一化的时候不是用类别归一化,而是用所有区域的分数进行归一化,通过区域之间的对比找到包含该类别信息最多的区域
- 某区域r属于某类别c的得分,为后两部分的积
- 全图的类别得分,为所有区域属于该类别的得分之和
训练的loss function如下
最后一项是一个校准项(按照理解轻微更改了,感觉论文notation有点问题),其目的是通过拉近feature的距离约束解的平滑性(即与正确解相近的proposal也应该得到高分)。
实验结果
本文根据basenet不同给出了4种model:S (VGG-F), M (VGG-M-1024), L (VGG-VD16)和Ens(前三种ensemble的模型)
- Ablation:
- Object proposal
- Baseline mAP: Selective Search S 31.1%, M 30.9%, L 24.3%, Ens. 33.3%
- Edge Box: +0~1.2%
- Edge Box + Edge Box Score: +1.8~5.9%
- Spatial regulariser (compared with Edge Box + Edge Box Score) mAP +1.2~4.4%
- Object proposal
- VOC2007
- mAP on test: S +2.9%, M +3.3%, L +3.2%, Ens. +7.7% compared with [36] + context
- CorLoc on trainval: S +5.7%, M +7.6%, L +5%, Ens. +9.5% compared with [36]
- Classification AP on test: S +7.9% compared with VGG-F, M +6.5% compared with VGG-M-1024, L +0.4% compared with VGG-VD16, Ens. -0.3% compared with VGG-VD16
- VOC2010
- mAP on test: +8.8% compared with [4]
- CorLoc on trainval: +4.5% compared with [4]
缺点
本文有一个明显的缺点是只考虑了一张图中某类别物体只出现一次的情况(regulariser中仅限制了最大值及其周围的框),这一点在文中给出的failure cases中也有所体现。
转载于:https://www.cnblogs.com/Xiaoyan-Li/p/8695772.html
[CVPR 2016] Weakly Supervised Deep Detection Networks论文笔记相关推荐
- 论文笔记 Weakly Supervised Deep Detection Networks - CVPR 2016
Weakly Supervised Deep Detection Networks Hakan Bilen, Andrea Vedaldi CVPR, 2016 (PDF) (Citations 58 ...
- 弱监督检测初识——Weakly Supervised Deep Detection Networks解读
文章目录 1 WSDDN模型 1.1 预训练CNN 1.2 CNN特征描述 1.2.1 区域推荐算法 1.2.2 Spatial partial pooling 1.2.3 推荐区域的特征描述 1.3 ...
- Procedural Noise Adversarial Examples for Black-Box Attacks on Deep Neural Networks论文笔记
Procedural Noise Adversarial Examples for Black-Box Attacks on Deep Neural Networks论文笔记 0. 概述 如今一些深度 ...
- Weighted-Entropy-based Quantization for Deep Neural Networks 论文笔记
---恢复内容开始--- 摘要 量化被认为是优化神经网络模型的推理成本的最有效方法之一,用于部署到具有严格资源限制的移动和嵌入式系统.在这种方法中,在严格的精度损失约束(例如,1%)下提供低成本量化是 ...
- 论文报告笔记(四) Towards Precise End-to-end Weakly Supervised Object Detection Network
文章目录 论文报告笔记(四) Towards Precise End-to-end Weakly Supervised Object Detection Network 弱监督学习 WSDDN OIC ...
- 论文笔记 Object-Aware Instance Labeling for Weakly Supervised Object Detection - ICCV 2019
Object-Aware Instance Labeling for Weakly Supervised Object Detection Kosugi ICCV, 2019 (PDF) (Citat ...
- Deep Anomaly Detection with Deviation Networks 论文笔记
Deep Anomaly Detection with Deviation Networks 论文笔记(自己回顾用) 问题描述 算法框架 伪代码 总结&个人计划 问题描述 论文利用少部分有标签 ...
- Utilizing the Instability in Weakly Supervised Object Detection (CVPR2019) 解读
论文链接:Utilizing the Instability in Weakly Supervised Object Detection 本文主要贡献: 通过分析检测器多示例学习(MIL)的不稳定性, ...
- 【模型压缩】Channel Pruning for Accelerating Very Deep Neural Networks算法笔记
转:https://blog.csdn.net/u014380165/article/details/79811779 论文:Channel Pruning for Accelerating Very ...
最新文章
- AI十年进阶:一段算法能力演进的历史
- Gitlab 官方对整个数据删除事件的详细说明
- VS2015 解决方案 或者项目 卡 正在加载 的解决办法
- .NET中的枚举(Enum)
- 枚举是如何实现的?(枚举的线程安全性及序列化问题)
- Atitit. 木马病毒的外部class自动加载机制------加载class的方法总结
- 查看SQL执行计划的方法及优劣
- 支付宝招“找茬”程序员,年薪无上限;谷歌宣布实现“量子霸权”;node.js 13.0.3 发布 | 极客头条...
- 求1-2+3-4+5 ... 99的所有数的和
- 《运算放大器权威指南》读书笔记(二)
- 获取全年所有非工作日和工作日的脚本(Python篇)
- iOS 获取汉字【简体中文】笔画数
- Excel 2010 VBA 入门 010 VBE编辑器的工具栏
- 【linux】按键盘Backspace键终端出现^H
- git两个分支代码合并
- 琴生Jensen不等式(期望意义下的形式)
- 谢谢 留下几个 那个有关于C++既QQ群吖
- 程序员薪酬高达 462 万元:是怎么回事?
- 魔法少女小圆计算机音乐,求魔法少女小圆一些背景音乐的名字
- 计算机文档打不开是什么原因,文档打不开电脑我的文档Word突然打不开了,是怎会是啊?其他还都 爱问知识人...
热门文章
- 民营企业SAP项目客户的几种心态
- 中文版GPT-3来了?智源、清华发布清源 CPM——以中文为核心的大规模预训练模型
- 最先进的目标检测器竟对大象“视而不见”
- 思科将在网络中融入人工智能和机器学习
- 李彦宏:汽车工业面临无人驾驶百年变局
- Python 之 Pandas (二)选择数据
- tar命令使用方法以及.tar.gz文件和.zip文件
- new_zeros()函数的作用
- 1000亿美元!英特尔要在美国建世界最大芯片厂,美520 亿美元芯片法案接近敲定...
- 多模态AI崛起,2022年人工智能5大发展趋势