小样本学习&元学习经典论文整理||持续更新

  本文整理了近些年来有关小样本学习的经典文章，并附上了原文下载链接以及论文解读链接。文末有我个人公众号“深视”的二维码链接，关注公众号回复“小样本学习”，可以打包下载全部文章。该文我会持续更新，不断增添新的文章和相关解读，大家可以收藏关注一下。

一、基于度量学习的小样本学习算法

1.《Siamese Neural Networks for One-shot Image Recognition》
  网络名称：Siamese Network
  文章来源：ICML2015
  原文下载：http://www.cs.toronto.edu/~gkoch/files/msc-thesis.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/105967091
2.《Matching Networks for One Shot Learning》
  网络名称：Matching Network
  文章来源：NIPS2016
  原文下载：https://arxiv.org/pdf/1606.04080.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/105908003
3.《Prototypical Networks for Few-shot Learning》
  网络名称：Prototypical Network
  文章来源：NIPS2017
  原文下载：https://arxiv.org/pdf/1703.05175.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/105894839
4.《Learning to Compare: Relation Network for Few-Shot Learning》
  网络名称：Relation Network
  文章来源：CVPR2018
  原文下载：https://arxiv.org/pdf/1711.06025.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106158012
5.《Finding Task-Relevant Features for Few-Shot Learning by Category Traversal》
  网络名称：CTM
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1905.11116.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106363521
6.《Variational Prototyping-Encoder: One-Shot Learning with Prototypical Images》
  网络名称：VPE
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1904.08482.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106438467
7.《RepMet: Representative-based metric learning for classification and few-shot object detection》
  网络名称：RepMet
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1806.04728.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106472082
8.《Revisiting Local Descriptor based Image-to-Class Measure for Few-shot Learning》
  网络名称：DN4
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1903.12290v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106479996
9.《Few-Shot Learning with Localization in Realistic Settings》
  网络名称：
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1904.08502v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106495421
10.《Dense Classification and Implanting for Few-Shot Learning》
  网络名称：
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1903.05050.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106545685
11.《TADAM: Task dependent adaptive metric for improved few-shot learning》
  网络名称：TADAM
  文章来源：NIPS2018
  原文下载：https://arxiv.org/pdf/1805.10123.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106641276
12.《Power Normalizing Second-order Similarity Network for Few-shot Learning》
  网络名称：SoSN
  文章来源：WACV2019
  原文下载：https://arxiv.org/abs/1811.04167v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106688690
13.《Few-Shot Learning with Metric-Agnostic Conditional Embeddings》
  网络名称：MACO
  文章来源：CVPR2018
  原文下载：https://arxiv.org/pdf/1802.04376v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106722822
14.《Improved Few-Shot Visual Classification》
  网络名称：Simple CNAPS
  文章来源：CVPR2020
  原文下载：https://arxiv.org/pdf/1912.03432.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106909418
15.《DeepEMD: Few-Shot Image Classification with Differentiable Earth Mover’s Distance and Structured Classifier》
  网络名称：DeepEMD
  文章来源：CVPR2020
  原文下载：https://arxiv.org/abs/2003.06777v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106942299
16.《Boosting Few-Shot Learning with Adaptive Margin Loss》
  网络名称：CRAML和TRAML
  文章来源：CVPR2020
  原文下载：https://arxiv.org/pdf/2005.13826.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106979488
17.《Adaptive Subspaces for Few-Shot Learning》
  网络名称：DSN
  文章来源：CVPR2020
  原文下载：http://openaccess.thecvf.com/content_CVPR_2020/papers/Simon_Adaptive_Subspaces_for_Few-Shot_Learning_CVPR_2020_paper.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106984460

二、基于参数优化的小样本学习算法

1.《Optimization as A Model for Few-shot Learning》
  网络名称：Meta-Learner LSTM
  文章来源：ICLR2017
  原文下载：https://openreview.net/pdf?id=rJY0-Kcll
  论文解读：https://blog.csdn.net/qq_36104364/article/details/105918760
2.《Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks》
  网络名称：MAML
  文章来源：ICML2017
  原文下载：https://arxiv.org/pdf/1703.03400.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/105982326
3.《Meta-SGD: Learning to Learn Quickly for Few-Shot Learning》
  网络名称：Meta-SGD
  文章来源：ICML2018
  原文下载：https://arxiv.org/pdf/1707.09835.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106096500
4.《Task-Agnostic Meta-Learning for Few-shot Learning》
  网络名称：TAML
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1805.07722.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106419974
5.《On First-Order Meta-Learning Algorithms》
  网络名称：Reptile
  文章来源：
  原文下载：https://arxiv.org/pdf/1803.02999v3.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106587228
6.《Deep Meta-Learning: Learning to Learn in the Concept Space》
  网络名称：DEML
  文章来源：华为诺亚方舟实验室
  原文下载：https://arxiv.org/abs/1802.03596.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106664598
7.《Meta-Learning of Neural Architectures for Few-Shot Learning》
  网络名称：MetaNAS
  文章来源：CVPR2020
  原文下载：https://arxiv.org/abs/1911.11090.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106922398
8.《Attentive Weights Generation for Few Shot Learning via Information Maximization》
  网络名称：AWGIM
  文章来源：CVPR2020
  原文下载：http://openaccess.thecvf.com/content_CVPR_2020/papers/Guo_Attentive_Weights_Generation_for_Few_Shot_Learning_via_Information_Maximization_CVPR_2020_paper.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/107007203

三、基于外部记忆的小样本学习算法

1.《Meta-Learning with Memory-Augmented Neural Networks》
  网络名称：MANN
  文章来源：ICML2016
  原文下载：https://arxiv.org/pdf/1605.06065v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106004802
2.《Meta Networks》
  网络名称：MetaNet
  文章来源：ICML2017
  原文下载：https://arxiv.org/pdf/1703.00837.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106197184
3.《Learning to remember rare events》
  网络名称：
  文章来源：ICLR2017
  原文下载：https://arxiv.org/pdf/1703.03129.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106266079
4.《Memory Matching Networks for One-Shot Image Recognition》
  网络名称：MM-Net
  文章来源：CVPR2018
  原文下载：https://arxiv.org/pdf/1804.08281.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106618750
5.《Dynamic Few-Shot Visual Learning without Forgetting》
  网络名称：
  文章来源：CVPR2018
  原文下载：https://arxiv.org/abs/1804.09458.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106627781

四、基于数据增强的小样本学习算法

1.《Low-Shot Visual Recognition by Shrinking and Hallucinating Features》
  网络名称：SGM
  文章来源：ICCV2017
  原文下载：https://arxiv.org/pdf/1606.02819v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106121741
2.《Meta-learning for semi-supervised few-shot classification》
  网络名称：
  文章来源：ICLR2018
  原文下载：https://arxiv.org/pdf/1906.00562.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106243619
3.《LaSO: Label-Set Operations networks for multi-label few-shot learning》
  网络名称：LaSONet
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1902.09811v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106333259
4.《Image Deformation Meta-Networks for One-Shot Learning》
  网络名称：IDeMe-Net
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1905.11641v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106355560
5.《Few-shot Learning via Saliency-guided Hallucination of Samples》
  网络名称：SalNet
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1904.03472v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106523254
6.《Low-Shot Learning from Imaginary Data》
  网络名称：PMN
  文章来源：CVPR2018
  原文下载：https://arxiv.org/pdf/1801.05401v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106590394
7.《Instance Credibility Inference for Few-Shot Learning》
  网络名称：ICI
  文章来源：CVPR2020
  原文下载：http://arxiv.org/abs/2003.11853.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106927937
8.《Adversarial Feature Hallucination Networks for Few-Shot Learning》
  网络名称：AFHN
  文章来源：CVPR2020
  原文下载：https://arxiv.org/pdf/2003.13193.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/107015984

五、基于语义信息的小样本学习算法

1.《Large-Scale Few-Shot Learning: Knowledge Transfer With Class Hierarchy》
  网络名称：
  文章来源：CVPR2019
  原文下载：https://www.researchgate.net/profile/Zhiwu_Lu2/publication/333602008_Large-Scale_Few-Shot_Learning_Knowledge_Transfer_With_Class_Hierarchy/links/5cf61bffa6fdcc847502e9de/Large-Scale-Few-Shot-Learning-Knowledge-Transfer-With-Class-Hierarchy.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106301958
2.《Generalized Zero- and Few-Shot Learning via Aligned Variational Autoencoders》
  网络名称：CADA-VAE
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1812.01784.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106381143
3.《TAFE-Net: Task-Aware Feature Embeddings for Low Shot Learning》
  网络名称：TAFE-Net
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1904.05967.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106412565
4.《Baby Steps Towards Few-Shot Learning with Multiple Semantics》
  网络名称：Multiple-Semantics
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1906.01905.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106566595
5.《Semantic Feature Augmentation in Few-shot Learning》
  网络名称：Dual-TriNet
  文章来源：ECCV2018
  原文下载：https://arxiv.org/pdf/1804.05298v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106714703
6.《Adaptive Cross-Modal Few-shot Learning》
  网络名称：AM3
  文章来源：NIPS2019
  原文下载：https://arxiv.org/pdf/1902.07104.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106734661

六、基于图神经网络的小样本学习算法

1.《Few-Shot Learning with Graph Neural Networks》
  网络名称：GNN
  文章来源：ICLR2018
  原文下载：https://arxiv.org/pdf/1711.04043.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106257218
2.《Generating Classification Weights with GNN Denoising Autoencoders for Few-Shot Learning》
  网络名称：DAE
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1905.01102v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106341245
3.《Edge-Labeling Graph Neural Network for Few-shot Learning》
  网络名称：EGNN
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1905.01436.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106504445
4.《DPGN: Distribution Propagation Graph Network for Few-shot Learning》
  网络名称：DPGN
  文章来源：CVPR2020
  原文下载：https://arxiv.org/pdf/2003.14247.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/107036021

七、基于深度强化学习的小样本学习算法

1.《Active One-shot Learning》
  网络名称：
  文章来源：NIPS2016
  原文下载：https://arxiv.org/pdf/1702.06559.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106141748
2.《Spot and Learn: A Maximum-Entropy Patch Sampler for Few-Shot Image Classification》
  网络名称：
  文章来源：CVPR2019
  原文下载：http://openaccess.thecvf.com/content_CVPR_2019/papers/Chu_Spot_and_Learn_A_Maximum-Entropy_Patch_Sampler_for_Few-Shot_Image_CVPR_2019_paper.pdf?source=post_page
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106428548

八、其他类型的小样本学习算法

1.《Meta-Learning with Temporal Convolutions》
  网络名称：TCML
  文章来源：ICLR2018
  原文下载：https://arxiv.org/pdf/1707.03141v2.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106235147
2.《Meta-Transfer Learning for Few-Shot Learning》
  网络名称：MTL
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1812.02391v2.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106403565
3.《Learning to propagate labels: Transductive propagation network for few-shot learning》
  网络名称：TPN
  文章来源：ICLR2019
  原文下载：http://arxiv.org/abs/1805.10002.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106670910
4.《Few-Shot Class-Incremental Learning》
  网络名称：TOPIC
  文章来源：CVPR2020
  原文下载：https://arxiv.org/pdf/2004.10956.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106855307
5.《Learning to Select Base Classes for Few-shot Classification》
  网络名称：
  文章来源：CVPR2020
  原文下载：https://arxiv.org/abs/2004.00315.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106956157
6.《TransMatch: A Transfer-Learning Scheme for Semi-Supervised Few-Shot Learning》
  网络名称：TransMatch
  文章来源：CVPR2020
  原文下载：https://arxiv.org/pdf/1912.09033.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/107022258

九、小样本语义分割算法

1.《One-Shot Learning for Semantic Segmentation》
  网络名称：
  文章来源：BMVC2017
  原文下载：https://arxiv.org/pdf/1709.03410.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106762117
2.《Conditional networks for few-shot semantic segmentation》
  网络名称：co-FCN
  文章来源：ICLR2018
  原文下载：https://openreview.net/pdf?id=SkMjFKJwG
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106766084
3.《CANet: Class-Agnostic Segmentation Networks with Iterative Refinement and Attentive Few-Shot Learning》
  网络名称：CANet
  文章来源：CVPR2019
  原文下载：https://arxiv.org/pdf/1903.02351.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106556797
4.《PANet: Few-Shot Image Semantic Segmentation with Prototype Alignment》
  网络名称：PANet
  文章来源：ICCV2019
  原文下载：https://arxiv.org/pdf/1908.06391.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106781918

十、小样本目标检测算法

1.《LSTD: A Low-Shot Transfer Detector for Object Detection》
  网络名称：LSTD
  文章来源：AAAI2018
  原文下载：https://arxiv.org/pdf/1803.01529.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106806507
2.《Few-Example Object Detection with Model Communication》
  网络名称：MSPLD
  文章来源：TPAMI2018
  原文下载：https://arxiv.org/pdf/1706.08249.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106824810
3.《Incremental Few-Shot Object Detection》
  网络名称：ONCE
  文章来源：CVPR2020
  原文下载：https://arxiv.org/pdf/2003.04668.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106838242
4.《Few-Shot Object Detection with Attention-RPN and Multi-Relation Detector》
  网络名称：
  文章来源：CVPR2020
  原文下载：https://arxiv.org/abs/1908.01998.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106873321
5.《Few-shot Object Detection via Feature Reweighting》
  网络名称：
  文章来源：ICCV2019
  原文下载：https://arxiv.org/abs/1812.01866.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106882520
6.《Meta R-CNN : Towards General Solver for Instance-level Low-shot Learning》
  网络名称：Meta R-CNN
  文章来源：ICCV2019
  原文下载：http://arxiv.org/abs/1909.13032v1.pdf
  论文解读：https://blog.csdn.net/qq_36104364/article/details/106886640
7.《》
  网络名称：
  文章来源：
  原文下载：
  论文解读：
如果大家对于深度学习与计算机视觉领域感兴趣，希望获得更多的知识分享与最新的论文解读，欢迎关注我的个人公众号“深视”。