检测

1. 单一物体检测

MTCNN:

Cascade-CNN: https://github.com/anson0910/CNN_face_detection

2. 通用物体检测

Faster-RCNN:

YOLO: https://github.com/pjreddie/darknet

SSD:

RetinaNet: https://github.com/fizyr/keras-retinanet

分类

VGG:

GoogLenet: https://github.com/n3011/Inception_v3_GoogLeNet

Resnet:

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

MNasNet:

1. 人脸识别

Deepface: https://github.com/RiweiChen/DeepFace

Normface:

Insightface: https://github.com/deepinsight/insightface

2. 文字识别

DeepOCR:

CTPN: https://github.com/tianzhi0549/CTPN (文字定位)

DenseNet + CTC:

YOLOv3 + CRNN : https://github.com/chineseocr/chineseocr

跟踪

1.

2.

分割

Unet:

mask-rcnn: https://github.com/matterport/Mask_RCNN

RCF:

图像复原

1. 去雨

DDN: https://github.com/XMU-smartdsp/Removing_Rain

CGAN:

DID-MDN: https://github.com/hezhangsprinter/DID-MDN

DeRaindrop:

2. 去雾

MSCNN: https://github.com/dishank-b/MSCNN-Dehazing-Tensorflow

DehazeNet:

3. 超分辨率

SRCNN: https://github.com/tegg89/SRCNN-Tensorflow

EDSR: (https://blog.csdn.net/xjp_xujiping/article/details/81986020)

4.图像单反化

DPED:

总结