gopala-kr / ConvNets

CNN Architectures

Back to top

• [LeNet(1998)] [paper]
• [LeNet-5 (2010)] [paper]
• [AlexNet (2012)] [paper]
• [ZFNet(2013)] [paper]
• [VGGNet (2014)] [paper]
• [GoogleNet/Inception(2014)] [paper]
• [FCN(2014)] [paper]
• [RCNN(2014)] [paper]
• [Deeply-supervised networks(2014)] [paper]
• [ResNet(2015)] [paper]
• [Ladder network(2015)] [paper]
• [YOLO(2015)] [Paper]
• [FractalNet (2016)] [paper]
• [PolyNet/Inception-Residual(2016)] [paper]
• [DenseNet(2016)] [paper] [code]
• [SegNet(2016)] [paper]
• [fast region based CNN(2016)] [paper]
• [Look up based CNN(2016)] [paper]
• [Deep network with stochastic depth(2016)] [paper]
• [ResNeXt(2016)] [paper]
• [SqueezeNet(2016)] [paper] [code]
• [CapsNet(2017)] [paper]
• [MobileNets(2017)] [paper]
• [Xception(2017)] [paper]
• [IRCNN(2017)][paper]
• [ViP CNN(2017)] [paper]
• [Squeeze-and-Excitation Networks(2017)][Paper] [code]
• [MobileFaceNets(2018)] [paper]
• [DCNet and DCNet++(2018)] [paper]

Applications

Back to top

• Object Recognition / Object Classification [SOTA]
• Object Detection [SOTA]
• Semantic Segmentation [SOTA]
• Object Tracking [SOTA]
• [Activity/Action Recognition] [SOTA]
• [Face Recognition]
• [Pose estimation]
• [Video & Image Captioning]
• [Biomedical Imaging] SOTA
• [Remote Sensing]
• [Video Analysis]
• [3D Vision]
• [CNNs for NLP]
• [CNNs for Speech Processing]
• [Adversarial Attacks on CNN] SOTA

Object Detection

Back to top

• [R-CNN(2013)] [paper]
• [Overfeat(2014)] [paper]
• [Multibox(2014)] [paper]
• [SPPNet(2014)] [paper]
• [MR-CNN(2015)] [paper]
• [Deepbox(2015)] [paper]
• [AttentionNet(2015)] [paper]
• [Fast R-CNN(2015)] [paper]
• [DeepProposal(2015)] [paper]
• [RPN(2015)] [[paper]]
• [Faster R-CNN(2015)] [paper]
• [YOLOv1(2016)] [paper]
• [GCNN(2016)] [paper]
• [AZNet(2016)] [paper]
• [ION(2016)] [paper]
• [HyperNet(2016)] [paper]
• [OHEM(2016)] [paper]
• [CRAFT(2016)] [paper]
• [MultipathNet(2016)] [paper]
• [SSD(2016)] [paper]
• [GBDNet(2016)] [paper]
• [CPF(2016)] [[paper]]
• [MS-CNN(2016)] [paper]
• [R-FCN(2016)] [paper]
• [PVANET(2016)] [paper]
• [DeepIDNet(2016)] [paper]
• [NoC(2016)] [paper]
• [DSSD(2017)] [paper]
• [TDM(2017)] [paper]
• [Feature Pyramid Net(2017)] [paper]
• [YOLOv2(2017)] [paper]
• [RON(2017)] [paper]
• [DCN(2017)] [[paper]]
• [DeNet(2017)] [paper]
• [CoupleNet(2017)] [paper]
• [RetinaNet(2017)] [paper]
• [Mask R-CNN(2017)] [paper]
• [DSOD(2017)] [paper]
• [SMN(2017)] [paper]
• [YOLOv3(2018)] [paper]
• [SIN(2018)] [paper]
• [STDN(2018)] [paper]
• [RefineDet(2018)] [paper]
• [RFBNet(2018)] [paper]

• [Light-Head R-CNN(2017)] [paper]
• [Cascade R-CNN(2017)] [paper]
• [YOLT(2018)] [paper]
• [FSSD(2018)] [paper]
• [ESSD] [paper]
• [MDSSD(2018)] [paper]
• [Pelee(2018)] [paper]
• [Fire SSD(2018)] [paper]
• [MegNet(2018)] [paper]
• [DetNet(2018)] [paper]
• [SSOD(2018)] [paper]
• [CornerNet(2018)] [paper]
• [3D Object Detection(2018)] [paper]
• [ZSD（Zero-Shot Object Detection）(2018)] [paper]
• [OSD（One-Shot object Detection）(2018)] [paper]
• [Weakly Supervised Object Detection(2018)] [paper]
• [Softer-NMS (2018)] [paper]
• [VideoCapsuleNet(2018)] [paper]
• [YOLO3D(2018)] [paper]

Semantic Segmentation

Back to top

• U-Net [arxiv][Pytorch]
• SegNet [arxiv][Caffe]
• DeepLab [arxiv][Caffe]
• FCN [arxiv][tensorflow]
• ENet [arxiv][Caffe]
• LinkNet [arxiv][Torch]
• DenseNet [arxiv[]
• Tiramisu [arxiv]
• DilatedNet [arxiv]
• PixelNet [arxiv][Caffe]
• ICNet [arxiv][Caffe]
• ERFNet [arxiv][Torch]
• RefineNet [arxiv][tensorflow]
• PSPNet [arxiv,pspnet][Caffe]
• Dilated convolution [arxiv][Caffe]
• DeconvNet [arxiv][Caffe]
• FRRN [arxiv][Lasagne]
• GCN [arxiv][PyTorch]
• LRR [arxiv][Matconvnet]
• DUC, HDC [arxiv][PyTorch]
• MultiNet [arxiv] [tensorflow1tensorflow2]
• Segaware [arxiv][Caffe]
• Semantic Segmentation using Adversarial Networks [arxiv] [Chainer]
• In-Place Activated BatchNorm:obtain #1 positions [arxiv] [Pytorch]
• [SegCaps(2018)] [arxiv]
• [SegFinNet(2018)] [arxiv]
• [SUNets(2018)] [arxiv]

References

• Computer Vision SOTA
• neural-network-papers
• ref-implementations
• ConvNet
• image-video-classification
• object-detection
• object-tracking-and-recognition
• semantic-segmentation
• image-generation
• human-pose-estimation
• low-level-vision
• vision-and-nlp
• other
• face-recognition
• cv-datasets

• First paper on Memorization in DNNs: https://arxiv.org/abs/1611.03530
• A closer look at memorization in Deep Networks: https://arxiv.org/abs/1706.05394
• Opening the Black Box of Deep Neural Networks via Information: https://arxiv.org/abs/1703.00810
• Understanding Deep Convolutional Networks
• A Survey: Time Travel in Deep Learning Space: An Introduction to Deep Learning Models and How Deep Learning Models Evolved from the Initial Ideas
• Survey on the attention based RNN model and its applications in computer vision
• Automatic Description Generation from Images: A Survey of Models, Datasets, and Evaluation Measures
• Best Practices for Applying Deep Learning to Novel Applications
• Adversarial Examples: Attacks and Defenses for Deep Learning
• Adversarial Examples that Fool both Human and Computer Vision
• Deep Learning: A Critical Appraisal
• A disciplined approach to neural network hyper-parameters: Part 1 -- learning rate, batch size, momentum, and weight decay
• Multi-objective Architecture Search for CNNs
• Deep Learning for Generic Object Detection: A Survey
• What I learned from competing against a ConvNet on ImageNet
• neural-network-architectures
• Real-time Object Detection with YOLO, YOLOv2 and now YOLOv3
• Recognition, Detection, Segmentation and Tracking
• Computer Vision: Algorithms and Applications
• Dynamical Isometry and a Mean Field Theory of CNNs: How to Train 10,000-Layer Vanilla Convolutional Neural Networks
• Adversarial Attacks and Defences: A Survey
• Secure Deep Learning Engineering: A Software Quality Assurance Perspective
• Face Recognition: From Traditional to Deep Learning Methods
• Security for Machine Learning-based Systems: Attacks and Challenges during Training and Inference
• Deep Learning For Computer Vision Tasks: A review
• Mini-batch Serialization: CNN Training with Inter-layer Data Reuse

Maintainer

Gopala KR / @gopala-kr