cv方面的博文收藏 -- 持续更新

（持续更新）

=====

论文的整理：

Image Generation: https://handong1587.github.io/deep_learning/2015/10/09/image-generation.html

Image Retrieval: https://handong1587.github.io/deep_learning/2015/10/09/image-retrieval.html

Image / Video Captioning: https://handong1587.github.io/deep_learning/2015/10/09/image-video-captioning.html

NLP: https://handong1587.github.io/deep_learning/2015/10/09/nlp.html

Object Detection: https://handong1587.github.io/deep_learning/2015/10/09/object-detection.html

OCR: https://handong1587.github.io/deep_learning/2015/10/09/ocr.html

=====

Tutorial

- ICML 2016 Tutorial on Deep Residual Networks (by Kaiming He): http://kaiminghe.com/icml16tutorial/index.html

- Generative Adversarial Networks (GANs): http://www.iangoodfellow.com/slides/2016-12-04-NIPS.pdf

- Welcome to PyTorch Tutorials: http://pytorch.org/tutorials/

=====

Code/Github:

- Faster-RCNN_TF: https://github.com/bigsnarfdude/Faster-RCNN_TF

- LSTM for Human Activity Recognition: https://github.com/guillaume-chevalier/LSTM-Human-Activity-Recognition/

- Fully Convolutional Instance-aware Semantic Segmentation: https://github.com/msracver/FCIS

- Deformable Convolutional Networks: https://github.com/msracver/Deformable-ConvNets

- MobileNet-Caffe: https://github.com/shicai/MobileNet-Caffe

=====

GAN

- AdversarialNetsPapers: https://github.com/zhangqianhui/AdversarialNetsPapers

- 从头开始GAN: https://zhuanlan.zhihu.com/p/27012520?group_id=870619199494565888

=====

博文的整理：

- 如何评价kaiming的Focal Loss for Dense Object Detection？https://www.zhihu.com/question/63581984

- 专栏 | Momenta详解ImageNet 2017夺冠架构SENet http://www.sohu.com/a/161633191_465975

- 如何评价Kaiming He最新的Mask R-CNN?: https://www.zhihu.com/question/57403701

- 如何评价 MSRA 最新的 Deformable Convolutional Networks？：https://www.zhihu.com/question/57493889

- Instance Segmentation 比 Semantic Segmentation 难很多吗？：https://www.zhihu.com/question/51704852/answer/142762733

- 如何评价周志华教授新提出的 Deep Forest 模型，它会取代当前火热的深度学习 DNN 吗？: https://www.zhihu.com/question/56474891

- Recurrent neural nets with Caffe: http://christopher5106.github.io/deep/learning/2016/06/07/recurrent-neural-net-with-Caffe.html

- 基于Lstm、CTC的验证码识别 使用mxnet与deep learning实现验证码识别: http://ranjun.me/2016/08/17/ocr-recaptcha/

- 端到端的OCR：验证码识别: https://zhuanlan.zhihu.com/p/21344595

- Machine learning algorithms: https://github.com/rushter/MLAlgorithms/blob/master/README.md

- 两分钟demo：快速Neural style艺术风格转换: http://phunter.farbox.com/post/mxnet-tutorial2

- Dilated Convolutions and Kronecker Factored Convolutions: http://www.inference.vc/dilated-convolutions-and-kronecker-factorisation/

深度学习变革视觉实例搜索 http://www.leiphone.com/news/201608/Iy4k4tU8zXThBYEQ.html

- MXnet初体验之inception-resnet-v2从Model到Predict http://hacker.duanshishi.com/?p=1733

- Neural Network Architectures: https://culurciello.github.io/tech/2016/06/04/nets.html

- 简述生成式对抗网络: http://chenrudan.github.io/blog/2016/11/12/gan.html

- 翻译《深度学习》: https://github.com/ExtremeMart/DeepLearningBook-CN

- 谷歌深度学习公开课任务 6: LSTMs：http://www.hankcs.com/ml/task-lstms-6.html

- Faster R-CNN Pedestrian and Car Detection: https://bigsnarf.wordpress.com/2016/11/07/faster-r-cnn-pedestrian-and-car-detection/

- Tensorflow: How to freeze a model and serve it with a python API: https://blog.metaflow.fr/tensorflow-how-to-freeze-a-model-and-serve-it-with-a-python-api-d4f3596b3adc#.i6zvxbm75

- Image Classification and Segmentation with Tensorflow and TF-Slim: http://warmspringwinds.github.io/tensorflow/tf-slim/2016/10/30/image-classification-and-segmentation-using-tensorflow-and-tf-slim/

- 用TensorFlow生成周杰伦歌词: http://leix.me/2016/11/28/tensorflow-lyrics-generation/

- Deep Learning for Detection of Diabetic Eye Disease: https://research.googleblog.com/2016/11/deep-learning-for-detection-of-diabetic.html

- Finding Beautiful Yelp Photos Using Deep Learning: https://engineeringblog.yelp.com/2016/11/finding-beautiful-yelp-photos-using-deep-learning.html

- 人群计数：Single-Image Crowd Counting via Multi-Column Convolutional Neural Network(CVPR2016): http://blog.csdn.net/u011285477/article/details/51954989

- Image Preprocessing with OpenCV: http://sujitpal.blogspot.hk/2016/11/image-preprocessing-with-opencv.html

- The major advancements in Deep Learning in 2016: https://tryolabs.com/blog/2016/12/06/major-advancements-deep-learning-2016/

- 研究|对偶学习：一种新的机器学习范式: http://weibo.com/ttarticle/p/show?id=2309404050111117306404#_0

- NIPS 2016 | Best Paper, Dual Learning, Review Network, VQA 等论文选介: http://mp.weixin.qq.com/s?__biz=MzAwMjM3MTc5OA==&mid=2652692311&idx=1&sn=c63a1ab15d61f2d0e4f85c9e941d44ec

- 卷积神经网络（CNN）在无人驾驶中的应用: http://geek.csdn.net/news/detail/129128

- OpenCV检测场景内是否有移动物体: http://www.tuicool.com/articles/2AnqeaN

- 机器学习之识别简单验证码: https://www.iswin.org/2016/10/15/Simple-CAPTCHA-Recognition-with-Machine-Learning/

- 深度学习在计算机视觉领域的前沿进展: https://zhuanlan.zhihu.com/p/24699780

- Sequence Classification with LSTM Recurrent Neural Networks in Python with Keras: http://machinelearningmastery.com/sequence-classification-lstm-recurrent-neural-networks-python-keras/

- AI技术讲座精选：神经结构搜索和强化学习: http://mp.weixin.qq.com/s?__biz=MzI0ODcxODk5OA==&mid=2247483966&idx=1&sn=e3fde0461e10e220aca322ca9395958c

- What is zero-shot learning?：https://www.quora.com/What-is-zero-shot-learning

- 【大规模图像检索的利器】Deep哈希算法介绍：https://zhuanlan.zhihu.com/p/21396173

- 请问CNN 中的 maxpool 到底是什么原理，为什么要取最大值，取最大值的原理是什么？谢谢: https://www.zhihu.com/question/41948919

- SVDNet for Pedestrian Retrieval：CNN 到底认为哪个投影方向是重要的？：https://zhuanlan.zhihu.com/p/29326061

======

NLP：

- 用Python实现文档聚类: http://python.jobbole.com/85481/

- DeepMind 机器理解文本 NLP 技术复现与解析: http://www.dlworld.cn/YuYiLiJie/2852.html

- GAN在自然语言处理方面有哪些有趣的文章和应用: https://www.zhihu.com/question/54463527

=====

标注工具

- 一个简单粗暴的人脸认证标注工具的实现: http://blog.jobbole.com/109684/

=====

数据集

医学影像开源数据集: https://zhuanlan.zhihu.com/p/24634505

=====

如果这篇博文对你有帮助，可否赏笔者喝杯奶茶？