![](https://img-blog.csdnimg.cn/20201014180756916.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
SSL
_Summer tree
这个作者很懒,什么都没留下…
展开
-
【PSMA】Progressive Sample Mining and Representation Learning for One-Shot Re-ID
文章目录主要挑战主要的贡献和创新点提出的方法总体框架与算法Vanilla pseudo label sampling (PLS)PLS with adversarial learningTraining losses实验与结果结论导言文章提出了一种新的三元组损失 HSoften-Triplet-Loss,在处理one-shot Re-ID任务中的噪声伪标签样本方面非常强大。文章还提出了一种伪标签采样过程,确保了在保持高可靠性的同时为训练图像形成正对和负对的可行性。与此同时,文章采用对抗学习网络,为训练原创 2020-12-02 18:40:59 · 509 阅读 · 0 评论 -
【VSA】One-shot video-based person re-identification with variance subsampling algorithm
文章目录解决了什么问题主要贡献和创新点基本框架提出的方法01 variance confidence方差置信度02 Variance Subsampling Algorithm 方差二次采样算法03 Variance decay strategy 方差衰减策略实验01 性能02 Ablation - sampling criterions导言针对现有工作中存在的错误伪标签问题,文章通过优化样本间的相似性度量和伪标签置信度评估策略来改善这个问题,从而提供模型性能。具体地,文章提出了方差置信度的概念,并原创 2020-11-27 21:11:34 · 365 阅读 · 0 评论 -
【MMT】ICLR 2020: MMT(Mutual Mean-Teaching)方法,无监督域适应在Person Re-ID上性能再创新高
为了减轻噪音伪标签的影响,文章提出了一种无监督的MMT(Mutual Mean-Teaching)方法,通过在迭代训练的方式中使用离线精炼硬伪标签和在线精炼软伪标签,来学习更佳的目标域中的特征。同时,还提出了可以让Traplet loss支持软标签的soft softmax-triplet loss”。 该方法在域自适应任务方面明显优于所有现有的Person re-ID方法,改进幅度高达18.2%。MUTUAL MEAN-TEACHING: PSEUDO LABEL REFINERY FOR UNSUP.原创 2020-07-07 19:58:11 · 1571 阅读 · 0 评论 -
【FSR】Feature Space Regularization for Person Re-Identification with One Sample
Feature Space Regularization for Person Re-Identification with One SampleAbstractI. INTRODUCTIONFramework.Our Method.II. RELATED WORKSA. Supervised Re-IDB. Semi-supervised Re-IDC. Unsupervised re-IDD. Progressive LearningIII. THE PROPOSED METHODA. Overall.原创 2020-07-05 17:36:15 · 526 阅读 · 4 评论 -
【CRR-FMM】A Concise Review of Recent Few-shot Meta-learning Methods
【CRR-FMM】A Concise Review of Recent Few-shot Meta-learning Methods1 IntroductionMindMap2. The Framework of Few-shot Meta-learning2.1. Notation and definitionsDefinition 1. (Small-sample learning)Definition 2. (Few-shot learning)Definition 3. (Few-shot met.原创 2020-05-29 19:16:30 · 481 阅读 · 0 评论 -
少标签数据学习:宾夕法尼亚大学Learning with Few Labeled Data
文章目录Few-shot image classificationThree regimes of image classificationProblem formulationA flavor of current few-shot algorithmsHow well does few-shot learning work today?The key ideaTransductive LearningAn exampleResults on benchmark datasetsThe ImageNet.原创 2020-05-27 10:42:05 · 837 阅读 · 0 评论 -
最新小样本学习综述 A Survey on Few-Shot Learning | 四大模型Multitask Learning、Embedding Learning、External Memory…
文章目录01 Multitask Learning01.1 Parameter Sharing01.2 Parameter Tying.02 Embedding Learning相关阅读:A Survey on Few-Shot Learning | Introduction and OverviewA Survey of Few-Shot Learing | Data给定少数样本的Dt...原创 2020-05-11 10:56:59 · 2536 阅读 · 1 评论 -
Generalizing from a Few Examples: A Survey on Few-Shot Learning 小样本学习最新综述 | 三大数据增强方法
文章目录01 Transforming Samples from Dtrain02 Transforming Samples from a Weakly Labeled or Unlabeled Data Set03 Transforming Samples from Similar Data SetsDiscussion and Summary上一篇:A Survey on Few-Shot ...原创 2020-04-29 15:53:18 · 3140 阅读 · 0 评论 -
Generalizing from a Few Examples: A Survey on Few-Shot Learning 小样本学习最新综述| Introduction and Overview
Author listYAQING WANG, Hong Kong University of Science and Technology and Baidu ResearchQUANMING YAO∗, 4Paradigm Inc.JAMES T. KWOK, Hong Kong University of Science and TechnologyLIONEL M. NI, Ho...原创 2020-04-13 22:21:16 · 2322 阅读 · 0 评论 -
Plant Leaves Classification: A Few-Shot Learning Method Based on Siamese Network
Plant Leaves Classification: A Few-Shot LearningMethod Based on Siamese NetworkAbstractIntroductionPROPOSED CNN STRUCTUREINITIAL CNN ANALYSISEXPERIMENTAL STRUCTURE AND ALGORITHMSAbstractIn this pap...原创 2019-11-24 19:35:29 · 17045 阅读 · 1 评论 -
Dynamic Label Graph Matching for Unsupervised Video Re-Identification
Dynamic Label Graph Matching for Unsupervised Video Re-IdentificationAbstractIntroductionAbstractThis pa-per focuses on cross-camera label estimation, which can be subsequently used in feature learn...原创 2019-10-22 11:19:14 · 1682 阅读 · 0 评论 -
Stepwise Metric Promotion for Unsupervised Video Person Re-identification
Stepwise Metric Promotion for Unsupervised Video Person Re-identificationAbstractIntroductionframeworkOur Method.ReferencesAbstracttwo assumptions two assumptionsdifferent video track-lets typical...原创 2019-10-21 21:28:17 · 940 阅读 · 0 评论 -
ZstGAN: An Adversarial Approach forUnsupervised Zero-Shot Image-to-Image Translation
ZstGAN: An Adversarial Approach forUnsupervised Zero-Shot Image-to-Image TranslationAbstractIntroductionMethodsProblem FormulationArchitectureReferenceAbstractIn this work In this workwe,we propose...原创 2019-10-24 10:44:54 · 1089 阅读 · 0 评论 -
Decoupled Novel Object Captioner
Decoupled Novel Object CaptionerAbstractIntroductionMethodsPreliminariesZero-Shot Novel Object Captioning.Sequence Model with the PlaceholderKey-Value Object MemoryFramework OverviewReference(image-t...原创 2019-10-24 21:42:52 · 1037 阅读 · 0 评论 -
A Survey of Zero-Shot Learning: Settings, Methods, and Applications [reading notes]
文章目录A Survey of Zero-Shot Learning: Settings, Methods, and ApplicationsIntroductionrestrictions(限制)Existing plansome popular application scenariosOverview of Zero-Shot Learningthe definition of zero-s...原创 2019-03-10 11:35:18 · 1730 阅读 · 0 评论 -
Single-Shot Refinement Neural Network for Object Detection[reading notes]
Single-Shot Refinement Neural Network for Object Detection文章目录Single-Shot Refinement Neural Network for Object Detectionquestion1:什么是two-stage approach?[参考链接](http://www.mamicode.com/info-detail-2305...原创 2019-03-22 11:33:34 · 399 阅读 · 0 评论 -
Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning[R]
Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning文章目录Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise LearningA...原创 2019-04-01 17:16:23 · 1016 阅读 · 0 评论 -
Progressive Learning for Person Re-Identification with One Example[reading notes]
# 导言> 这篇文章是Yu Wu在继[Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning(EUG)](https://blog.csdn.net/NGUever15/article/details/88930864)之后发表的一篇文章,文中工作基于EUG进行,对其进行进行了优化。建议阅读本文之前先阅读EUG对工作内容有初步了解,本博文也针对Progressive原创 2019-04-18 15:18:24 · 1635 阅读 · 3 评论 -
基于少量样本的快速学习Few-shot learning
基于少量样本的快速学习Few-shot learning 背景人工智能神经网络的三次浪潮深度学习人工智能困境人工智能 → 人类智能定义及数值原理机器学习定义数值原理数据增强数据预处理综合运用其他已有数据网络模型和相关任务共同训练——多任务学习模型生成模型嵌入学习模型(Embedding Learning)带外部存储的学习模型塑性网络模型贝叶斯程序学习(BPL)模型优化算法研究展望参考文献特别鸣谢...原创 2019-08-08 09:51:07 · 1974 阅读 · 0 评论 -
Zero-Shot Deep Domain Adaptation[reading notes]
文章目录AbstractIntroductionAbstractDomain adaptation(域适应) is an important tool to transfer knowledge about a task.Current approaches:假设 task-relevant target-domain数据在训练期间是可用的。而我们展示了如何在 上述数据不可用的情况下实现...原创 2019-08-24 15:04:05 · 1329 阅读 · 1 评论 -
Zero-shot Learning Using Synthesised Unseen Visual Data with Diffusion Regularisation [reading note]
文章目录Zero-shot Learning Using Synthesised Unseen Visual Data with Diffusion RegularisationIntroduction and Ralation worksZero-shot Recognition SchemesSemantic Side InformationStructure-Preserving Proje...原创 2019-03-09 17:04:48 · 475 阅读 · 0 评论