Paper Reading

까먹으면 다시 보려고 정리합니다.

Classification

• Lenet-5(1998), PyTorch Code [Google Colab / Blog Posting]

• AlexNet(2012), PyTorch Code [Google Colab / Blog Posting]

• Pseudo Label(2013)

• PyTorch 구현 코드로 살펴보는 Knowledge Distillation(2014), PyTorch Code [Google Colab / Blog Posting], paper [pdf]

• GoogLeNet(2014), PyTorch Code [Google Colab / Blog Posting]

• VGGNet(2014), PyTorch Code [Google Colab / Blog Posting]

• Inception-v3(2015)

• ResNet(2015), PyTorch Code [Google Colab / Blog Posting]

• Pre-Activation ResNet(2016), PyTorch Code [Google Colab / Blog Posting]

• WRN, Wide Residual Networks(2016), PyTorch Code [Google Colab / Blog Posting]

• SqueezeNet(2016)

• Inception-v4(2016), PyTorch Code [Google Colab / Blog Posting]

• PyramidNet(2017)

• DenseNet(2017), PyTorch Code [Google Colab / Blog Posting]

• Xception(2017), PyTorch Code [Google Colab / Blog Posting]

• MobileNetV1(2017), PyTorch Code [Google Colab / Blog Posting]

• ResNext(2017), PyTorch Code [Google Colab / Blog Posting]

• PolyNet(2017)

• Residual Attention Network(2017), PyTorch Code [Google Colab / Blog Posting]

• DPN(2017)

• Non-local Neural Network(2017), paper [pdf]

• SENet(2018), PyTorch Code [Google Colab / Blog Posting]

• MobileNetV2(2018)

• ShuffleNet(2018)

• NasNet(2018)

• PNasNet(2018)

• ShuffleNet(2018)

• CondenseNet(2018)

• CBAM(2018), paper [pdf]

• Bag of Tricks(2019)

• MobileNetV3(2019)

• EfficientNet(2019), PyTorch Code [Google Colab / Blog Posting]

• SKNet(2019), paper [pdf]

• BiT(2019). paper [pdf]

• Noisy Student(2020)

• FixEfficientNet(2020)

• Meta Pseudo Labels(2020)

• Noise or Signal: The Role of Image Backgrounds in Object Recognition(2020), paper [pdf]

• VIT(2020), paper [pdf], PyTorch Code [Google Colab / Blog Posting]

• Deit(2020), paper [pdf]

• EfficientNetV2(2021)

• Knowledge distillation: A good teacher is patient and consitent(2021), paper [pdf]

• MLP-Mixer(2021), paper [odf]

• CvT(2021), paper [pdf]

• CeiT(2021), paper [pdf]

• Early Convolutions Help Transformers See Better(2021), paper [pdf]

• BoTNet(2021), paper [pdf]

• Conformer(2021), paper [pdf]

• Delving Deep into the Generalization of Vision Transformers under Distribution Shifts(2021), paper [pdf]

• Scaling Vision Transformers(2021), paper [pdf]

• CMT(2021), paper [pdf]

Object Detection

• IoU(Intersection over Union)를 이해하고 파이토치로 구현하기

• 비-최대 억제(NMS, Non-maximum Suppression)를 이해하고 파이토치로 구현하기

• mAP(mean Average Precision)을 이해하고 파이토치로 구현하기

• R-CNN(2013)

• SPPnet(2014)

• Fast R-CNN(2014)

• Faster R-CNN(2015) 미완성 PyTorch Code

• SSD(2016)

• YOLO v1(2016)

• R-FCN(2016)

• OHEM(2016)

• DSSD(2017)

• YOLO v2(2017)

• FPN(2017)

• RetinaNet(2017) PyTorch Code [Google Colab / Blog Posting]

• RON(2017)

• DCN(2017)

• CoupleNet(2017)

• Soft-NMS(2017)

• RefineDet(2018)

• Cascade R-CNN(2018)

• YOLO v3(2018), PyTorch Code [Google Colab / Blog Posting]

• CornerNet(2018)

• M2Det(2019)

• CenterNet(2019), paper [pdf]

• Gaussian YOLOv3(2019), paper [pdf]

• FCOS(2019), paper [pdf]

• YOLOv4(2020), paper [pdf]

• EfficientDet(2020), paper [pdf]

• CSPNet(2020), paper [pdf]

• DIoU Loss(2020), paper [pdf], Code

• CircleNet(2020), paper [pdf]

• DETR(2020), paper [pdf]

• ACT(2020), paper [pdf]

• Deformable DETR(2020), paper [pdf]

• Localization Distillation for Dense Object Detection(2102)

• CenterNet2(2021), paper [pdf]

• Swin Transformer(2021), paper [pdf]

• YOLOr(2021), paper [pdf]

• YOLOS(2021), paper [pdf]

• Dynamic Head, Unifying Object Detection Heads with Attention(2021), paper [pdf]

• Pix2Seq(2021), paper [pdf]

• Anchor DETR, Query Design for Transformer-Based Object Detection(2021), paper [pdf]

• DAB-DETR, Dynamic Anchor Boxes are Better Queries for DETR(2022), paper [pdf]

• DN-DETR, Accelerate DETR Training by Introducing Query DeNoising(2022), paper [pdf]

• DINO, DETR with Imporved DeNoising Anchor Boxes for End-to-End Object Detection(2022), paper [pdf]

Segmentation

• DeepLabV1(2014)

• FCN(2015)

• DeConvNet(2015)

• DilatedNet(2015), paper [pdf]

• PyTorch 구현 코드로 살펴보는 SegNet(2015), paper [pdf]

• PSPNet(2016), paper [pdf]

• DeepLabv3(2017), paper [pdf]

• Mask R-CNN(2017)

• PANet(2018), paper [pdf]

• Panoptic Segmentation(2018), paper [pdf]

• Weakly- and Semi-Supervised Panoptic Segmentation(2018), paper [pdf]

• Panoptic Segmentation with a Joint Semantic and Instance Segmentation Network(2018), paper [pdf]

• Single Network Panoptic Segmentation for Street Scene Understanding(2019), paper [pdf]

• Panoptic Feature Pyramid Networks(2019), paper [pdf]

• IMP: Instance Mask Projection for High Accuracy Semantic Segmentation of Things(2019), paper [pdf]

• Object-Contextual Representations for Semantic Segmentation(2019), paper [pdf]

• CondInst, Conditional Convolution for Instance Segmentation(2020), paper [pdf]

• Max-DeepLab, End-to-End Panoptic Segmentation wtih Mask Transformers, paper [pdf]

• MaskFormer, Per-Pixel Classification is Not All You Need for Semantic Segmentation(2021), paper [pdf]

• Open-World Entity Segmentation(2021), paper [pdf]

• Prompt based Multi-modal Image Segmentation(2021), paper [pdf]

• DenseCLIP, Language-Guided Dense Prediction with Context-Aware Prompting, paper [pdf]

• Mask2Former, Masked-attention Mask Transformer for Universal Image Segmentation(2021)

• SeMask<, Semantically Masked Transformers for Semantic Segmentation(2021)

Self-supervised Learning

• Constrative Loss(2006), paper [pdf]

• Exemplar-CNN(2014), paper [pdf]

• Unsupervised Learning of Visual Representation using Videos, paper [pdf]

• Context Prediction(2015), paper [pdf]

• Jigsaw Puzzles(2016), paper [odf]

• Colorful Image Coloriztion(2016), paper [pdf]

• Deep InfoMax(2018), paper [pdf]

• Deep Cluster(2018), paper [pdf]

• IIC(2018), paper [pdf]

• Rotation(2018), paper [pdf]

• Unsupervised Feature Learning via Non-Parametric Instance Discrimination(2018), paper [pdf]

• ADMIN(2019), paper [pdf]

• Contrastive Multiview Coding(2019), paper [pdf]

• MoCo(2019), paper [pdf]

• SeLa(2019), paper [pdf]

• SimCLR(2020), paper [pdf]

• MoCov2(2020), PyTorch Code [Google Colab / Blog Posting], paper [pdf]

• SimSiam(2020), paper [pdf]

• Understanding the Behaviour of Contrastive Loss(2020), paper [pdf]

• BYOL(2020), paper [pdf]

• SwAV(2020), paper [pdf]

• PCL(2020), paper [pdf]

• SimCLRv2(2020), paper [pdf]

• Supervised Contrastive Learning(2020), paper [pdf]

• DenseCL(2020), Dense Contrastive Learning for Self-Supervised Visual Pre-Training, paper [pdf]

• DetCo(2021), paper [pdf

• SCRL(2021), paper [pdf]

• MoCov3(2021), paper [pdf]

• DINO(2021), paper [pdf]

• EsViT(2021), paper [pdf]

• Masked Autoencoders Are Scalable Vision Learners(2021), paper [pdf]

Video SSL

• Tracking Emerges by Colorizing Videos(2018), paper [pdf]

• Self-supervised Learning for Video Correspondence Flow(2019), paper [pdf]

• Learning Correspondence from the Cycle-consistency of Time(2019), paper [pdf]

• Joint-task Self-supervised Learning for Temporal Correspondence(2019), paper [pdf]

• MAST(2020), https://arxiv.org/abs/2002.07793

• Space-Time Correspondence as a Contrastive Random Walk(2020), paper [pdf]

• Contrastive Transformation for Self-supervised Correspondence Learning(2020), paper [pdf]

• Mining Better Samples for Contrastive Learning of Temporal Correspondence(2021), paper [pdf]

• VFS(2021), paper [pdf]

• Contrastive Learning of Image Representations with Cross-Video Cycle-Consistency, paper [pdf]

• ViCC(2021), paper [pdf]

Semi-supervised Learning

• Temporal ensembling for semi-supervised learning(2016) , paper [pdf]

• Mean teachers are better role models(2017), paper [pdf]

• Consistency-based Semi-supervised Learning for Object Detection(2019), paper [pdf]

• PseudoSeg, Designing Pseudo Labels for Semantic Segmentation(2020), paper [pdf]

• ReCo, Bootstrapping Semantic Segmentation with Regional Contrast(2021), paper [pdf]

• Semi-Supervised Semantic Segmentation with Cross Pseudo Supervision(2021), paper [pdf]

• Soft Teacher(2021), End-to-End Semi-Supervised Object Detection with Soft Teacher, paper [pdf]

• CaSP(2021), Class-agnostic Semi-Supervised Pretraining for Detection & Segmentation, paper [pdf]

weakly

• Class Activation Map(CAM), Learning Deep Features for Discriminative Localization, paper [pdf]

• Grad-CAM, Visual Explanations from Deep Networks via Gradient based Localization, paper [pdf]

• Zoom-CAM, Generating Fine-grained Pixel Annotations from Image Labels(2020), paper [pdf]

• GETAM: Gradient-weighted Element-wise Transformer Attention Map for Weakly-supervised Semantic Segmentation(2021), paper [pdf]

Video Recognition

• Learning Spatiotemporal Features with 3D Convolutional Network(2014), paper [pdf]

• Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset(2017), paper [pdf]

• SlowFast Networks for Video Recognition(2018), paper [pdf]

• TSM(2018), paper [pdf]

• GCNet(2019), paper [pdf]

• Drop an Octave(2019), paper [pdf]

• STM(2019), paper [pdf]

• X3D(2020), paper [pdf]

• VTN(2021). paper [pdf]

• TimeSformer(2021), paper [pdf], Youtube [link]

• ViViT(2021), paper [pdf]

• MViT(2021), paper [pdf]

• X-ViT(2021), paper [pdf]

• Video Swin Transformer(2021), paper [pdf]

• Towards Training Stronger Video Vision Transformers for EPIC-KITCHENS-100 Action Recognition(2021), paper [pdf]

• VLF(2021), paper [pdf], Youtube [link]

Video Segmentation

• VisTR(2020), paper [pdf]

Zero Shot Classification

• DeViSE, A Deep Visual-Semantic Embedding Model(2013), paper [pdf]

• Zero-shot Learning via Shared-Reconstruction-Graph Pursuit(2017), paper [pdf]

• A Generative Adversarial Approach for Zero-Shot Learning from Noisy Texts(2017), paper [pdf]

• f-VAEGAN-D2, A Feature Generating Framework for Any Shot Learning(2019), paper [pdf]

• TCN(2019), Transferable Contrastive Network for Generalized Zero-Shot Learning, paper [pdf]

• Rethinking Zero-Shot Learning: A Conditional Visual Classification Perspective(2019), paper [pdf]

• Convolutional Prototype Learning for Zero-Shot Recognition(2019), paper [pdf]

• DRN, Class-Prototype Discriminative Network for Generalized Zero-Shot Learning(2020), paper [pdf]

• DAZLE(2020), Fine-Grained Generalized Zero-Shot Learning via Dense Attribute-Based Attention, paper [pdf]

• IPN(2021), Isometric Propagation Network for Generalized Zero-Shot Learning, paper [pdf]

• CE-GZSL(2021), Contrastive Embedding for Generalized Zero-Shot Learning, paper [pdf]

• Task-Independent Knowledge Makes for Transferable Represenatations for Generalized Zero-Shot Learning(2021), paper [pdf]

• Zero-Shot Learning via Contrastive Learning on Dual Knowledge Graphs(2021), paper [pdf]

• FREE: Feature Refinement for Generalized Zero-Shot Learning(2021), paper [pdf]

• ALIGN(2021), Scaling Up Visual and Vision-Language Representation Learning with Noisy Text Supervision, paper [pdf]

• LiT: Zero-Shot Transfer with Locked-image Text Tuning(2021), paper [pdf]

• Generalized Category Discovery(2022), paper [pdf]

Zero Shot Detection

• Synthesizing the Unseen for Zero-shot Object Detection(2020), paper [pdf]

• ViLD(2021), Open-Vocabulary Object Detection via Vision and Language Knowledge Distillation, paper [pdf]

• Robust Region Feature Synthesizer for Zero-Shot Object Detection(2022), paper [pdf]

• Detic(2022), Detecting Twenty-thousand Classes using Image-level Supervision, paper [pdf]

Zero Shot Segmentation

• Zero-Shot Semantic Segmentation(2019), paper [pdf]

• Semantic Projection Network for Zero- and Few-Label Semantic Segmentation(2020), paper [pdf]

• Learning unbiased zero-shot semantic segmentation networks via transductive transfer(2020), paper [pdf]

• A review of Generalized Zero-Shot Learning Methods(2020), paper [pdf]

• Consistent Structural Relation Learning for Zero-Shot Segmentation(2020, paper [pdf]

• Uncertainty-Aware Learning for Zero-Shot Semantic Segmentation(2020), paper [pdf]

• Context-aware Feature Generation for Zero-shot Semantic Segmentation(2020), paper [pdf]

• Recursive Training for Zero-Shot Semantic Segmentation(2021), paper [pdf]

• Zero-Shot Instance Segmentation(2021), paper [pdf]

• A Closer Look at Self-training for Zero-Label Segmantic Segmentation(2021), paper [pdf]

• Prototypical Matching and Open Seg Rejection for Zero-Shot Semantic Segmentation(2021), paper [pdf]

• SIGN(2021), Spatial-information Incorporated Generative Network for GGeneralized Zero-shot Semantic Segmentation, paper [pdf]

• Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation(2021), paper [pdf]

• Languager Driven Semantic Segmentation(2021), paper [pdf]

• Zero-Shot Semantic Segmentation via Spatial and Multi-Scale Aware Visual Class Embedding, paper [pdf]

• DenseCLIP: Extract Free Dence Labels from CLIP(2021), paper [pdf]

• Decoupling Zero-Shot Semantic Segmentation(2021), paper [pdf]

• A Simple Baseline for Zero-Shot Semantic Segmentation with Pre-trained Vision-language Model(2021), paper [pdf]

• Open-Vocabulary Image Segmentation(2021), paper [pdf]

Few-Shot, Meta Learning

• Matching Networks for One Shot Learning(2016), paper [pdf]

• Learning to Compare: Relation Network for Few-Shot Learning(2017), paper [pdf]

Prompting and Vision-Language Model

cv

• CPT, Colorful Prompt Tuning for Pre-trained Vision-Language Models

• CoOp(2021), Learning to Prompt for Vision-Language Models

• CLIP-Adapter, Better Vision-Language Models with Feature Adapters(2021)

• Tip-Adapter, Training-free CLIP-Adapter for Better Vision-Language Modeling

• Prompt-Based Multi-Model Image Segmentation(2021)

• DenseCLIP, Language-Guided Dense Prediction with Context-Aware Prompting(2021)

• Prompting Visual-Language Models for Efficient Video Understanding, paper [pdf]

• Conditianl Prompt Learning for Visiona-Language Models, paper [pdf]

nlp

• Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing(2021)

Image Processing

• PyTorch 구현 코드로 살펴보는 SRCNNe(2014), PyTorch Code [Google Colab / Blog Posting], paper [pdf]

• FlowNet(2015), paper [pdf]

• PWC-Net(2017), paper [pdf]

• Residual Non-local Attention Networks for Image Restoration(2019), paper [pdf]

3D Vision

• Convolutional-Recursive Deep Learning for 3D Object Classification(2012), paper [pdf]

• PointNet(2016), paper [pdf]

• Set Transformer(2018), paper [pdf]

• Centroid Transformer(2021), paper [pdf]

NLP

• PyTorch 코드로 살펴보는 Seq2Seq(2014), PyTorch Code [Google Colab / Blog Posting], paper [pdf]

• PyTorch 코드로 살펴보는 GRU(2014), paper [pdf]

• PyTorch 코드로 살펴보는 Attention(2015), paper [odf]

• PyTorch 코드로 살펴보는 Convolutional Sequence to Sequence Learning(2017), paper [pdf]

• PyTorch 코드로 살펴보는 Transforemr(2017), paper [pdf]

• BERT(2018), paper [pdf]

• ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators(2020), paper [pdf]

GAN

• PyTorch 구현 코드로 살펴보는 GAN(2014), PyTorch Code [Google Colab / Blog Posting], paper [pdf]

• PyTorch 구현 코드로 살펴보는 CGAN(2014), PyTorch Code [Google Colab / Blog Posting], paper [pdf]

• Generative Moment Matching Network(2015), paper [pdf]

• PyTorch 구현 코드로 살펴보는 DCGAN(2015), PyTorch Code [Google Colab / Blog Posting], paper [pdf]

• PyTorch 구현 코드로 살펴보는 Pix2Pix(2016), PyTorch Code [Google Colab / Blog Posting], paper [pdf]

Active Learning

• Towards Reducing Labeling Cost in Deep Object Detection(2021), paper [pdf]

Pose estimation

• Hourglass(2016)

long tail

• Class-Balanced Loss(2019), paper [pdf]

• Seesaw Loss for Long-Tailed Instance Segmentation(2020), paper [pdf]

Face Recognition

• Pytorch 구현 코드로 살펴보는 FaceNet(2015), paper [pdf]

Model Compression

• Deep Compression(2016), paper [pdf]

Activation Function

• Mish(2019), paper [pdf]

Augmentation

• CutMix(2019), paper [pdf]

• Learning Data Augmentation Strategies for Object Detection(2019, paper [pdf]

• Simple Copy-Paste is a Strong Data Augmentation Method for Instance Segmentation(2020), paper [pdf]

Style Transfer

• PyTorch 구현 코드로 살펴보는 A Neural Algorithm of Artistic Style(2016), PyTorch Code [Google Colab / Blog Posting], paper [pdf]

Regularization

• DropBlock(2018), paper [pdf]

Normalization

• Batch Normalization(2015)

• Group Normalization(2018), paper [pdf]

• Cross iteration BN(2020), paper [pdf]

Optimization

• An overview of gradient descent optimization algorithm(2017)

• AdamW(2017), paper [pdf]