深度學習有開原始碼的文獻
人群分析
Novel Dataset for Fine-grained Abnormal Behavior Understanding in Crowd
人群異常行為資料庫：https://github.com/hosseinm/med Panic，Fight，Congestion，Obstacle ，Neutral
Flexible Spatio-Temporal Networks for Video Prediction CVPR2017
影象語義匹配
SCNet: Learning Semantic Correspondence ICCV2017
Matlab code:
特徵匹配
GMS: Grid-based Motion Statistics for Fast, Ultra-robust Feature Correspondence CVPR2017
c++ code: https://github.com/JiawangBian/GMS-Feature-Matcher
目標檢測 — 加速候選區域提取
DeNet: Scalable Real-time Object Detection with Directed Sparse Sampling ICCV2017
https://github.com/lachlants/denet
殘差網路超快訓練
Super-Convergence: Very Fast Training of Residual Networks Using Large Learning Rates
https://github.com/lnsmith54/super-convergence
深度邊緣檢測
Richer Convolutional Features for Edge Detection CVPR2017
https://github.com/yun-liu/rcf
目標檢測
A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection ECCV2016
目標檢測
RON: Reverse Connection with Objectness Prior Networks for Object Detection CVPR2017
https://github.com/taokong/RON
同時檢測和分割，類似 Mask R-CNN
BlitzNet: A Real-Time Deep Network for Scene Understanding ICCV2017
https://github.com/dvornikita/blitznet
網路裁剪加速
Learning Efficient Convolutional Networks through Network Slimming ICCV2017
https://github.com/liuzhuang13/slimming
CNN網路通道裁剪加速
Channel Pruning for Accelerating Very Deep Neural Networks ICCV2017
https://github.com/yihui-he/channel-pruning
目標候選區域分割
FastMask: Segment Multi-scale Object Candidates in One Shot CVPR2017
https://github.com/voidrank/FastMask
新的池化方法
S3Pool: Pooling with Stochastic Spatial Sampling CVPR2017
https://github.com/Shuangfei/s3pool
CNN 影象檢索
Efficient Diffusion on Region Manifolds: Recovering Small Objects with Compact CNN Representations CVPR2017
https://github.com/ahmetius/diffusion-retrieval
基於查表的CNN網路模型
LCNN: Lookup-based Convolutional Neural Network CVPR2017
https://github.com/hessamb/lcnn
資料擴張 Image Augmentation tool
Augmentor: An Image Augmentation Library for Machine Learning
GitHub: https://github.com/mdbloice/Augmentor
人臉識別
SphereFace: Deep Hypersphere Embedding for Face Recognition CVPR2017
https://github.com/wy1iu/sphereface
語義分割
The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation CVPRW 2017
Theano/Lasagne code Implementation
https://github.com/0bserver07/One-Hundred-Layers-Tiramisu
語義分割
Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes CVPR2017
https://github.com/TobyPDE/FRRN
目標檢測
DSOD: Learning Deeply Supervised Object Detectors from Scratch ICCV2017
https://github.com/szq0214/DSOD
DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
https://bitbucket.org/aquariusjay/deeplab-public-ver2/overview
Fully Convolutional Instance-aware Semantic Segmentation CVPR2017
https://github.com/msracver/FCIS
SqueezeDet: Unified, Small, Low Power Fully Convolutional Neural Networks
for Real-Time Object Detection for Autonomous Driving
https://github.com/BichenWuUCB/squeezeDet
Network Dissection:
Quantifying Interpretability of Deep Visual Representations
CVPR2017
https://github.com/CSAILVision/NetDissect
A Discriminative Feature Learning Approach for Deep Face Recognition
code: https://github.com/ydwen/caffe-face
目標檢測：
PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection
https://github.com/sanghoon/pva-faster-rcnn
R-FCN: Object Detection via Region-based Fully Convolutional Networks
https://github.com/daijifeng001/r-fcn
A-Fast-RCNN: Hard Positive Generation via Adversary for Object Detection CVPR 2017
Caffe code : https://github.com/xiaolonw/adversarial-frcnn
行人檢測：
深度去噪
Learning Deep CNN Denoiser Prior for Image Restoration
https://github.com/cszn/ircnn