Learning a discriminative mid level feature for action recognition

一种基于TensorFlow实现的新的深度学习图像识别技术。

Gradient-Based Learning Applied to Document Recognition 极好的文！尤其是p5-p9，一定要细看，反复琢磨。把LeNet-5的来龙去脉讲得清清楚楚

MachineLearningandFeatureEngineeringforComputerNetworkSecurityJonathanJ.Davis

Coupled Dictionary and Feature Space Learning with Applications to cross-domain image synthesis and

2019年IEEEAccess上发表的一篇行人再识别文章。本文提出一个多分支深度模型。具体地，这个模型包括5个分支，这5个分支中，两个分支从水平和垂直方向学习带有空间信息的部分特征；一个分支学习由骨干

Omni-Scale Feature Learning for Person Re-Identification论文预训练权重

基于上下文的编码器：图像修复的特征学习 Wepresentanunsupervisedvisualfeaturelearningalgo-rithmdrivenbycontext-basedpixe

Heterogeneous Multi-task Semantic Feature Learning forClassification

Tracking feature extraction based on manifold learning framework

A Feature Extraction Method Based on Dictionary Learning for EEG