Basic Knowledge

An overview of gradient descent optimization algorithms: arxiv

Classification

AlexNet: paper
VGG: arxiv
Inception: v1 v2 v3 v4 & Inception-ResNet
ResNet: v1 v2
RexNeXt: arxiv
DenseNet: arxiv

LightWeight Network

MobileNet: v1 v2
SqueezeNet: arxiv
ShuffleNet: v1 v2
Xception: arxiv
IGCV2: arxiv

Object Detection

R-CNN: arxiv pycaffe
Fast R-CNN: arxiv pycaffe
Faster R-CNN: arxiv pycaffe matcaffe

PVANet: arxiv code

SSD: arxiv caffe
DSSD: arxiv MXNet

YOLO: v1 v2 v3 darknet

R-FCN: arxiv code
FPN: arxiv
Deformable ConvNets: arxiv
RFBNet: arxiv
MegDet: arxiv

Segmentation

FCN: arxiv
Mask R-CNN: arxiv
DeconvNet: arxiv
SegNet: arxiv

Network Analysis

Speed accuracy trade-offs for modern convolutional object detectors: arxiv
Residual Networks Behave Like Ensembles of Relatively Shallow Networks: arxiv

Network Training

Focal Loss: arxiv
OHEM: arxiv