xiaoyufenfei/awesome-computer-vision-models

It's the list with popular deep learning models related to classification and segmentation task

awesome-computer-vision-models

This is a list with popular classification and segmentation models with corresponding evaluation metrics.

You can check some of the models using tensorflow.js demo application.

Classification models

• AlexNet ('One weird trick for parallelizing convolutional neural networks') [2014]
• VGG/BN-VGG ('Very Deep Convolutional Networks for Large-Scale Image Recognition') [2014]
• ResNet ('Deep Residual Learning for Image Recognition') [2015]
• InceptionV3 ('Rethinking the Inception Architecture for Computer Vision') [2015]
• PreResNet ('Identity Mappings in Deep Residual Networks') [2016]
• DenseNet ('Densely Connected Convolutional Networks') [2016]
• PyramidNet ('Deep Pyramidal Residual Networks') [2016]
• ResNeXt ('Aggregated Residual Transformations for Deep Neural Networks') [2016]
• WRN ('Wide Residual Networks') [2016]
• Xception ('Xception: Deep Learning with Depthwise Separable Convolutions') [2016]
• InceptionV4/InceptionResNetV2 ('Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning') [2016]
• PolyNet ('PolyNet: A Pursuit of Structural Diversity in Very Deep Networks') [2016]
• DarkNet ('Darknet: Open source neural networks in C') [2016?]
• ResAttNet ('Residual Attention Network for Image Classification') [2017]
• CondenseNet ('CondenseNet: An Efficient DenseNet using Learned Group Convolutions') [2017]
• DRN-C/DRN-D ('Dilated Residual Networks') [2017]
• DPN ('Dual Path Networks') [2017]
• ShuffleNet ('ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices') [2017]
• DiracNetV2 ('DiracNets: Training Very Deep Neural Networks Without Skip-Connections') [2017]]
• SENet/SE-ResNet/SE-PreResNet/SE-ResNeXt ('Squeeze-and-Excitation Networks') [2017]
• MobileNet ('MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications') [2017]
• NASNet ('Learning Transferable Architectures for Scalable Image Recognition') [2017]
• DLA ('Deep Layer Aggregation') [2017]
• AirNet/AirNeXt ('Attention Inspiring Receptive-Fields Network for Learning Invariant Representations') [2018]
• BAM-ResNet ('BAM: Bottleneck Attention Module') [2018]
• CBAM-ResNet ('CBAM: Convolutional Block Attention Module') [2018]
• SqueezeNet/SqueezeResNet ('SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size') [2016]
• SqueezeNext ('SqueezeNext: Hardware-Aware Neural Network Design') [2018]
• ShuffleNetV2 ('ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design') [2018]
• MENet ('Merging and Evolution: Improving Convolutional Neural Networks for Mobile Applications') [2018]
• FD-MobileNet ('FD-MobileNet: Improved MobileNet with A Fast Downsampling Strategy') [2018]
• MobileNetV2 ('MobileNetV2: Inverted Residuals and Linear Bottlenecks') [2018]
• IGCV3 ('IGCV3: Interleaved Low-Rank Group Convolutions for Efficient Deep Neural Networks') [2018]
• DARTS ('DARTS: Differentiable Architecture Search') [2018]
• PNASNet ('Progressive Neural Architecture Search') [2018]
• Amoeba ('Regularized Evolution for Image Classifier Architecture Search') [2018]
• MnasNet ('MnasNet: Platform-Aware Neural Architecture Search for Mobile') [2018]
• IBN-Net ('Two at Once: Enhancing Learning andGeneralization Capacities via IBN-Net') [2018]
• MarginNet ('Large Margin Deep Networks for Classification') [2018]
• A^2 Nets ('A^2-Nets: Double Attention Networks') [2018]
• FishNet ('FishNet: A Versatile Backbone for Image, Region, and Pixel Level Prediction') [2018]
• Shape-ResNet ('IMAGENET-TRAINED CNNS ARE BIASED TOWARDS TEXTURE; INCREASING SHAPE BIAS IMPROVES ACCURACY AND ROBUSTNESS')[2019]
• Shift-Invariant-ResNet ('Making Convolutional Networks Shift-Invariant Again')[2019]
• SimCNN ('Greedy Layerwise Learning Can Scale to ImageNet')[2019]
• SKNet ('Selective Kernel Networks')[2019]
• SRM ('SRM : A Style-based Recalibration Module for Convolutional Neural Networks')[2019]
• EfficientNet ('EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks')[2019]
• ProxylessNAS ('PROXYLESSNAS: DIRECT NEURAL ARCHITECTURE SEARCH ON TARGET TASK AND HARDWARE')[2019]
• MixNet ('MixNet: Mixed Depthwise Convolutional Kernels')[2019]
• ECA-Net ('ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks')[2019]
• ACNet-Densenet121 ('ACNet: Strengthening the Kernel Skeletons for Powerful CNN via Asymmetric Convolution Blocks')[2019]
• LIP-* ('LIP: Local Importance-based Pooling')[2019]
• MuffNet ('MuffNet: Multi-Layer Feature Federation for Mobile Deep Learning')[2019]
• *-Bin-5 ('Making Convolutional Networks Shift-Invariant Again')[2019]
• FixRes ResNeXt-101 WSL ('Fixing the train-test resolution discrepancy')[2019]
• Noisy Student (L2) (using extra-data) ('Self-training with Noisy Student improves ImageNet classification')[2019]

Model	Number of parameters	FLOPS	Top-1 Error	Top-5 Error	DEMO
AlexNet	62.3M	1,132.33M	40.96	18.24	X
VGG-16	138.3M	?	26.78	8.69	X
ResNet-10	5.5M	894.04M	34.69	14.36	Try live
ResNet-18	11.7M	1,820.41M	28.53	9.82	Try live
ResNet-34	21.8M	3,672.68M	24.84	7.80	Try live
ResNet-50	25.5M	3,877.95M	22.28	6.33	Try live
Inception v3	23.8M	?	21.2	5.6	X
PreResNet-18	11.7M	1,820.56M	28.43	9.72	Try live
PreResNet-34	21.8M	3,672.83M	24.89	7.74	Try live
PreResNet-50	25.6M	3,875.44M	22.40	6.47	Try live
DenseNet-121	8.0M	2,872.13M	23.48	7.04	Try live
DenseNet-161	28.7M	7,793.16M	22.86	6.44	X
PyramidNet-101	42.5M	8,743.54M	21.98	6.20	X
ResNeXt-14(32x4d)	9.5M	1,603.46M	30.32	11.46	Try live
ResNeXt-26(32x4d)	15.4M	2,488.07M	24.14	7.46	Try live
WRN-50-2	68.9M	11,405.42M	22.53	6.41	X
Xception	22,855,952	8,403.63M	20.97	5.49	X
InceptionV4	42,679,816	12,304.93M	20.64	5.29	X
InceptionResNetV2	55,843,464	13,188.64M	19.93	4.90	X
PolyNet	95,366,600	34,821.34M	19.10	4.52	X
InceptionResNetV2	55,843,464	13,188.64M	19.93	4.90	X
DarkNet Ref	7,319,416	367.59M	38.58	17.18	Try live
DarkNet Tiny	1,042,104	500.85M	40.74	17.84	Try live
DarkNet 53	41,609,928	7,133.86M	21.75	5.64	Try live
Attention-92	51.3M	?	19.5	4.8	X
CondenseNet (G=C=8)	4.8M	?	26.2	8.3	X
DPN-68	12,611,602	2,351.84M	23.24	6.79	Try live
ShuffleNet x1.0 (g=1)	1,531,936	148.13M	34.93	13.89	Try live
DiracNetV2-18	11,511,784	1,796.62M	31.47	11.70	Try live
DiracNetV2-34	21,616,232	3,646.93M	28.75	9.93	Try live
SENet-16	31,366,168	5,081.30M	25.65	8.20	Try live
SENet-154	115,088,984	20,745.78M	18.62	4.61	X
MobileNet x1.0	4,231,976	579.80M	26.61	8.95	Try live
NASNet-A 4@1056	5,289,978	584.90M	25.68	8.16	Try live
NASNet-A 6@4032	88,753,150	23,976.44M	18.14	4.21	X
DLA-34	15,742,104	3,071.37M	25.36	7.94	Try live
AirNet50-1x64d (r=2)	27.43M	?	22.48	6.21	X
BAM-ResNet-50	25.92M	?	23.68	6.96	X
CBAM-ResNet-50	28.1M	?	23.02	6.38	X
SqueezeResNet1.1	1,235,496	352.02M	40.09	18.21	Try live
SqueezeNet1.1	1,235,496	352.02M	39.31	17.72	Try live
1.0-SqNxt-23v5	921,816	285.82M	40.77	17.85	X
1.5-SqNxt-23v5	1,953,616	550.97M	33.81	13.01	X
2.0-SqNxt-23v5	3,366,344	897.60M	29.63	10.66	X
ShuffleNetV2 x1.0	2,278,604	149.72M	31.44	11.63	Try live
456-MENet-24×1(g=3)	5.3M	?	28.4	9.8	X
FD-MobileNet x1.0	2,901,288	147.46M	34.23	13.38	Try live
MobileNetV2 x1.0	3,504,960	329.36M	26.97	8.87	Try live
IGCV3	3.5M	?	28.22	9.54	X
DARTS	4.9M	?	26.9	9.0	X
PNASNet-5	5.1M	?	25.8	8.1	X
AmoebaNet-C	5.1M	?	24.3	7.6	X
MnasNet	4,308,816	317.67M	31.58	11.74	Try live
IBN-Net50-a	?	?	22.54	6.32	X
MarginNet	?	?	22.0	?	X
A^2 Net	?	?	23.0	6.5	X
FishNeXt-150	26.2M	?	21.5	?	X
Shape-ResNet	25.5M	?	23.28	6.72	X
ResNet-50-Bin-5	?	?	23.0	?	X
SimCNN(k=3 train)	?	?	28.4	10.2	X
SKNet-50	27.5M	?	20.79	?	X
SRM-ResNet-50	25.62M	?	22.87	6.49	X
EfficientNet-B0	5,288,548	414.31M	24.77	7.52	Try live
EfficientNet-B7b	66,347,960	39,010.98M	15.94	3.22	X
ProxylessNAS	?	?	24.9	7.5	X
MixNet-L	7.3M	?	21.1	5.8	X
ECA-Net50	24.37M	3.86G	22.52	6.32	X
ECA-Net101	7.3M	7.35G	21.35	5.66	X
ACNet-Densenet121	?	?	24.18	7.23	X
LIP-ResNet-50	23.9M	5.33G	21.81	6.04	X
LIP-ResNet-101	42.9M	9.06G	20.67	5.40	X
LIP-DenseNet-BC-121	8.7M	4.13G	23.36	6.84	X
MuffNet_1.0	2.3M	146M	30.1	?	X
MuffNet_1.5	3.4M	300M	26.9	?	X
ResNet-34-Bin-5	21.8M	3,672.68M	25.80	?	X
ResNet-50-Bin-5	25.5M	3,877.95M	22.96	?	X
MobileNetV2-Bin-5	3,504,960	329.36M	27.50	?	X
FixRes ResNeXt101 WSL	829M	?	13.6	2.0	X
Noisy Student*(L2)	480M	?	12.6	1.8	X

Segmentation models

Semantic segmentation

• U-Net ('U-Net: Convolutional Networks for Biomedical Image Segmentation') [2015]
• DeconvNet ('Learning Deconvolution Network for Semantic Segmentation') [2015]
• ParseNet ('ParseNet: Looking Wider to See Better') [2015]
• Piecewise ('Efficient piecewise training of deep structured models for semantic segmentation') [2015]
• SegNet ('SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation') [2016]
• FCN ('Fully Convolutional Networks for Semantic Segmentation') [2016]
• ENet ('ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation') [2016]
• DilatedNet ('MULTI-SCALE CONTEXT AGGREGATION BY DILATED CONVOLUTIONS') [2016]
• PixelNet ('PixelNet: Towards a General Pixel-Level Architecture') [2016]
• RefineNet ('RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation') [2016]
• LRR ('Laplacian Pyramid Reconstruction and Refinement for Semantic Segmentation') [2016]
• FRRN ('Full-Resolution Residual Networks for Semantic Segmentation in Street Scenes') [2016]
• Semantic Segmentation using Adversarial Networks ('Semantic Segmentation using Adversarial Networks') [2016]
• MultiNet ('MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving') [2016]
• DeepLab ('DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs') [2017]
• LinkNet ('LinkNet: Exploiting Encoder Representations for Efficient Semantic Segmentation') [2017]
• Tiramisu ('The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation') [2017]
• ICNet ('ICNet for Real-Time Semantic Segmentation on High-Resolution Images') [2017]
• ERFNet ('Efficient ConvNet for Real-time Semantic Segmentation') [2017]
• PSPNet ('Pyramid Scene Parsing Network') [2017]
• GCN ('Large Kernel Matters — Improve Semantic Segmentation by Global Convolutional Network') [2017]
• Segaware ('Segmentation-Aware Convolutional Networks Using Local Attention Masks') [2017]
• PixelDCN ('PIXEL DECONVOLUTIONAL NETWORKS') [2017]
• DeepLabv3 ('Rethinking Atrous Convolution for Semantic Image Segmentation') [2017]
• DUC, HDC ('Understanding Convolution for Semantic Segmentation') [2018]
• ShuffleSeg ('SHUFFLESEG: REAL-TIME SEMANTIC SEGMENTATION NETWORK') [2018]
• AdaptSegNet ('Learning to Adapt Structured Output Space for Semantic Segmentation') [2018]
• TuSimple-DUC ('Understanding Convolution for Semantic Segmentation') [2018]
• R2U-Net ('Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image Segmentation') [2018]
• Attention U-Net ('Attention U-Net: Learning Where to Look for the Pancreas') [2018]
• DANet ('Dual Attention Network for Scene Segmentation') [2018]
• ENCNet ('Context Encoding for Semantic Segmentation') [2018]
• ShelfNet ('ShelfNet for Real-time Semantic Segmentation') [2018]
• LadderNet ('LADDERNET: MULTI-PATH NETWORKS BASED ON U-NET FOR MEDICAL IMAGE SEGMENTATION') [2018]
• ССС ('Concentrated-Comprehensive Convolutions for lightweight semantic segmentation') [2018]
• DifNet ('DifNet: Semantic Segmentation by Diffusion Networks') [2018]
• BiSeNet ('BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation') [2018]
• ESPNet ('ESPNet: Efficient Spatial Pyramid of Dilated Convolutions for Semantic Segmentation') [2018]
• SPADE ('Semantic Image Synthesis with Spatially-Adaptive Normalization') [2019]
• SeamlessSeg ('Seamless Scene Segmentation') [2019]
• EMANet ('Expectation-Maximization Attention Networks for Semantic Segmentation') [2019]

Model	PASCAL-Context	Cityscapes (mIOU)	PASCAL VOC 2012 (mIOU)	COCO Stuff	ADE20K VAL (mIOU)
U-Net	?	?	?	?	?
DeconvNet	?	?	72.5	?	?
ParseNet	40.4	?	69.8	?	?
Piecewise	43.3	71.6	78.0	?	?
SegNet	?	56.1	?	?	?
FCN	37.8	65.3	62.2	22.7	29.39
ENet	?	58.3	?	?	?
DilatedNet	?	?	67.6	?	32.31
PixelNet	?	?	69.8	?	?
RefineNet	47.3	73.6	83.4	33.6	40.70
LRR	?	71.8	79.3	?	?
FRRN	?	71.8	?	?	?
MultiNet	?	?	?	?	?
DeepLab	45.7	64.8	79.7	?	?
LinkNet	?	?	?	?	?
Tiramisu	?	?	?	?	?
ICNet	?	70.6	?	?	?
ERFNet	?	68.0	?	?	?
PSPNet	47.8	80.2	85.4	?	44.94
GCN	?	76.9	82.2	?	?
Segaware	?	?	69.0	?	?
PixelDCN	?	?	73.0	?	?
DeepLabv3	?	?	85.7	?	?
DUC, HDC	?	77.1	?	?	?
ShuffleSeg	?	59.3	?	?	?
AdaptSegNet	?	46.7	?	?	?
TuSimple-DUC	80.1	?	83.1	?	?
R2U-Net	?	?	?	?	?
Attention U-Net	?	?	?	?	?
DANet	52.6	81.5	?	39.7	?
ENCNet	51.7	75.8	85.9	?	44.65
ShelfNet	48.4	75.8	84.2	?	?
LadderNet	?	?	?	?	?
CCC-ERFnet	?	69.01	?	?	?
DifNet-101	45.1	?	73.2	?	?
BiSeNet(Res18)	?	?	74.7	28.1	?
ESPNet	?	?	63.01	?	?
SPADE	?	62.3	?	37.4	38.5
SeamlessSeg	?	77.5	?	?	?
EMANet	?	?	88.2	39.9	?

Detection models

• [R-CNN] Rich feature hierarchies for accurate object detection and semantic segmentation | Ross Girshick, Jeff Donahue, Trevor Darrell, Jitendra Malik | [CVPR' 14] |[pdf] [official code - caffe] [2014]
• [OverFeat] OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks | Pierre Sermanet, et al. | [ICLR' 14] |[pdf] [official code - torch] [2014]
• [MultiBox] Scalable Object Detection using Deep Neural Networks | Dumitru Erhan, et al. | [CVPR' 14] |[pdf] [2014]
• [SPP-Net] Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition | Kaiming He, et al. | [ECCV' 14] |[pdf] [official code - caffe] [unofficial code - keras] [unofficial code - tensorflow] [2014]
• [MR-CNN] Object detection via a multi-region & semantic segmentation-aware CNN model | Spyros Gidaris, Nikos Komodakis | [ICCV' 15] |[pdf] [official code - caffe] [2015]
• [DeepBox] DeepBox: Learning Objectness with Convolutional Networks | Weicheng Kuo, Bharath Hariharan, Jitendra Malik | [ICCV' 15] |[pdf] [official code - caffe] [2015]
• [AttentionNet] AttentionNet: Aggregating Weak Directions for Accurate Object Detection | Donggeun Yoo, et al. | [ICCV' 15] |[pdf] [2015]
• [Fast R-CNN] Fast R-CNN | Ross Girshick | [ICCV' 15] |[pdf] [official code - caffe] [2015]
• [DeepProposal] DeepProposal: Hunting Objects by Cascading Deep Convolutional Layers | Amir Ghodrati, et al. | [ICCV' 15] |[pdf] [official code - matconvnet] [2015]
• [Faster R-CNN, RPN] Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks | Shaoqing Ren, et al. | [NIPS' 15] |[pdf] [official code - caffe] [unofficial code - tensorflow] [unofficial code - pytorch]
• [YOLO v1] You Only Look Once: Unified, Real-Time Object Detection | Joseph Redmon, et al. | [CVPR' 16] |[pdf] [official code - c] [2016]
• [G-CNN] G-CNN: an Iterative Grid Based Object Detector | Mahyar Najibi, et al. | [CVPR' 16] |[pdf] [2016]
• [AZNet] Adaptive Object Detection Using Adjacency and Zoom Prediction | Yongxi Lu, Tara Javidi. | [CVPR' 16] |[pdf] [2016]
• [ION] Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks | Sean Bell, et al. | [CVPR' 16] |[pdf] [2016]
• [HyperNet] HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection | Tao Kong, et al. | [CVPR' 16] |[pdf] [2016]
• [OHEM] Training Region-based Object Detectors with Online Hard Example Mining | Abhinav Shrivastava, et al. | [CVPR' 16] |[pdf] [official code - caffe] [2016]
• [CRAPF] CRAFT Objects from Images | Bin Yang, et al. | [CVPR' 16] |[pdf] [official code - caffe] [2016]
• [MPN] A MultiPath Network for Object Detection | Sergey Zagoruyko, et al. | [BMVC' 16] |[pdf] [official code - torch] [2016]
• [SSD] SSD: Single Shot MultiBox Detector | Wei Liu, et al. | [ECCV' 16] |[pdf] [official code - caffe] [unofficial code - tensorflow] [unofficial code - pytorch] [2016]
• [GBDNet] Crafting GBD-Net for Object Detection | Xingyu Zeng, et al. | [ECCV' 16] |[pdf] [official code - caffe] [2016]
• [CPF] Contextual Priming and Feedback for Faster R-CNN | Abhinav Shrivastava and Abhinav Gupta | [ECCV' 16] |[pdf] [2016]
• [MS-CNN] A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection | Zhaowei Cai, et al. | [ECCV' 16] |[pdf] [official code - caffe] [2016]
• [R-FCN] R-FCN: Object Detection via Region-based Fully Convolutional Networks | Jifeng Dai, et al. | [NIPS' 16] |[pdf] [official code - caffe] [unofficial code - caffe] [2016]
• [PVANET] PVANET: Deep but Lightweight Neural Networks for Real-time Object Detection | Kye-Hyeon Kim, et al. | [NIPSW' 16] |[pdf] [official code - caffe] [2016]
• [DeepID-Net] DeepID-Net: Deformable Deep Convolutional Neural Networks for Object Detection | Wanli Ouyang, et al. | [PAMI' 16] |[pdf] [2016]
• [NoC] Object Detection Networks on Convolutional Feature Maps | Shaoqing Ren, et al. | [TPAMI' 16] |[pdf]
• [DSSD] DSSD : Deconvolutional Single Shot Detector | Cheng-Yang Fu1, et al. | [arXiv' 17] |[pdf] [official code - caffe] [2017]
• [TDM] Beyond Skip Connections: Top-Down Modulation for Object Detection | Abhinav Shrivastava, et al. | [CVPR' 17] |[pdf] [2017]
• [FPN] Feature Pyramid Networks for Object Detection | Tsung-Yi Lin, et al. | [CVPR' 17] |[pdf] [unofficial code - caffe] [2017]
• [YOLO v2] YOLO9000: Better, Faster, Stronger | Joseph Redmon, Ali Farhadi | [CVPR' 17] |[pdf] [official code - c] [unofficial code - caffe] [unofficial code - tensorflow] [unofficial code - tensorflow] [unofficial code - pytorch] [2017]
• [RON] RON: Reverse Connection with Objectness Prior Networks for Object Detection | Tao Kong, et al. | [CVPR' 17] |[pdf] [official code - caffe] [unofficial code - tensorflow] [2017]
• [DCN] Deformable Convolutional Networks | Jifeng Dai, et al. | [ICCV' 17] |[pdf] [official code - mxnet] [unofficial code - tensorflow] [unofficial code - pytorch] [2017]
• [DeNet] DeNet: Scalable Real-time Object Detection with Directed Sparse Sampling | Lachlan Tychsen-Smith, Lars Petersson | [ICCV' 17] |[pdf] [official code - theano] [2017]
• [CoupleNet] CoupleNet: Coupling Global Structure with Local Parts for Object Detection | Yousong Zhu, et al. | [ICCV' 17] |[pdf] [official code - caffe] [2017]
• [RetinaNet] Focal Loss for Dense Object Detection | Tsung-Yi Lin, et al. | [ICCV' 17] |[pdf] [official code - keras] [unofficial code - pytorch] [unofficial code - mxnet] [unofficial code - tensorflow] [2017]
• [Mask R-CNN] Mask R-CNN | Kaiming He, et al. | [ICCV' 17] |[pdf] [official code - caffe2] [unofficial code - tensorflow] [unofficial code - tensorflow] [unofficial code - pytorch] [2017]
• [DSOD] DSOD: Learning Deeply Supervised Object Detectors from Scratch | Zhiqiang Shen, et al. | [ICCV' 17] |[pdf] [official code - caffe] [unofficial code - pytorch] [2017]
• [SMN] Spatial Memory for Context Reasoning in Object Detection | Xinlei Chen, Abhinav Gupta | [ICCV' 17] |[pdf] [2017]
• [YOLO v3] YOLOv3: An Incremental Improvement | Joseph Redmon, Ali Farhadi | [arXiv' 18] |[pdf] [official code - c] [unofficial code - pytorch] [unofficial code - pytorch] [unofficial code - keras] [unofficial code - tensorflow] [2018]
• [ZIP] Zoom Out-and-In Network with Recursive Training for Object Proposal | Hongyang Li, et al. | [IJCV' 18] |[pdf] [official code - caffe] [2018]
• [SIN] Structure Inference Net: Object Detection Using Scene-Level Context and Instance-Level Relationships | Yong Liu, et al. | [CVPR' 18] |[pdf] [official code - tensorflow] [2018]
• [STDN] Scale-Transferrable Object Detection | Peng Zhou, et al. | [CVPR' 18] |[pdf]
• [RefineDet] Single-Shot Refinement Neural Network for Object Detection | Shifeng Zhang, et al. | [CVPR' 18] |[pdf] [official code - caffe] [unofficial code - chainer] [unofficial code - pytorch] [2018]
• [MegDet] MegDet: A Large Mini-Batch Object Detector | Chao Peng, et al. | [CVPR' 18] |[pdf] [2018]
• [DA Faster R-CNN] Domain Adaptive Faster R-CNN for Object Detection in the Wild | Yuhua Chen, et al. | [CVPR' 18] |[pdf] [official code - caffe] [2018]
• [SNIP] An Analysis of Scale Invariance in Object Detection – SNIP | Bharat Singh, Larry S. Davis | [CVPR' 18] |[pdf] [2018]
• [Relation-Network] Relation Networks for Object Detection | Han Hu, et al. | [CVPR' 18] |[pdf] [official code - mxnet] [2018]
• [Cascade R-CNN] Cascade R-CNN: Delving into High Quality Object Detection | Zhaowei Cai, et al. | [CVPR' 18] |[pdf] [official code - caffe] [2018]
• Finding Tiny Faces in the Wild with Generative Adversarial Network | Yancheng Bai, et al. | [CVPR' 18] |[pdf] [2018]
• [STDnet] STDnet: A ConvNet for Small Target Detection | Brais Bosquet, et al. | [BMVC' 18] |[pdf] [2018]
• [RFBNet] Receptive Field Block Net for Accurate and Fast Object Detection | Songtao Liu, et al. | [ECCV' 18] |[pdf] [official code - pytorch] [2018]
• Zero-Annotation Object Detection with Web Knowledge Transfer | Qingyi Tao, et al. | [ECCV' 18] |[pdf] [2018]
• [CornerNet] CornerNet: Detecting Objects as Paired Keypoints | Hei Law, et al. | [ECCV' 18] |[pdf] [official code - pytorch] [2018]
• [Pelee] Pelee: A Real-Time Object Detection System on Mobile Devices | Jun Wang, et al. | [NIPS' 18] |[pdf] [official code - caffe] [2018]
• [HKRM] Hybrid Knowledge Routed Modules for Large-scale Object Detection | ChenHan Jiang, et al. | [NIPS' 18] |[pdf] [2018]
• [MetaAnchor] MetaAnchor: Learning to Detect Objects with Customized Anchors | Tong Yang, et al. | [NIPS' 18] |[pdf] [2018]
• [M2Det] M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network | Jun Wang, et al. | [AAAI' 19] |[pdf] [2018]
• [Libra RetinaNet] Libra R-CNN: Towards Balanced Learning for Object Detection | Jiangmiao Pang, et al. | [pdf] [2019]
• [YOLACT-700] YOLACT Real-time Instance Segmentation | [pdf] [2019]
• [DetNASNet] DetNAS: Backbone Search for Object Detection | [pdf] [2019]

Detector	VOC07 (mAP@IoU=0.5)	VOC12 (mAP@IoU=0.5)	COCO (mAP)
R-CNN	58.5	-	-
OverFeat	-	-	-
MultiBox	29.0	-	-
SPP-Net	59.2	-	-
MR-CNN	78.2	73.9	-
AttentionNet	-	-	-
Fast R-CNN	70.0	68.4	-
Faster R-CNN	73.2	70.4	36.8
YOLO v1	66.4	57.9	-
G-CNN	66.8	66.4	-
AZNet	70.4	-	22.3
ION	80.1	77.9	33.1
HyperNet	76.3	71.4	-
OHEM	78.9	76.3	22.4
MPN	-	-	33.2
SSD	76.8	74.9	31.2
GBDNet	77.2	-	27.0
CPF	76.4	72.6	-
MS-CNN	-	-	-
R-FCN	79.5	77.6	29.9
PVANET	-	-	-
DeepID-Net	69.0	-	-
NoC	71.6	68.8	27.2
DSSD	81.5	80.0	-
TDM	-	-	37.3
FPN	-	-	36.2
YOLO v2	78.6	73.4	21.6
RON	77.6	75.4	-
DCN	-	-	-
DeNet	77.1	73.9	33.8
CoupleNet	82.7	80.4	34.4
RetinaNet	-	-	39.1
Mask R-CNN	-	-	39.8
DSOD	77.7	76.3	-
SMN	70.0	-	-
YOLO v3	-	-	33.0
SIN	76.0	73.1	23.2
STDN	80.9	-	-
RefineDet	83.8	83.5	41.8
MegDet	-	-	-
RFBNet	82.2	-	-
CornerNet	-	-	42.1
LibraRetinaNet	-	-	43.0
YOLACT-700	-	-	31.2
DetNASNet(3.8)	-	-	42.0