smkim7-kr/Deep-Learning-paper-study

Reviews/Codes for DL Research Papers

Deep Learning Paper Review

• Classic Papers

  • Deep Residual Learning for Image Recognition :: Pytorch / Tensorflow2
  • You Only Look Once: Unified, Real-Time Object Detection
  • EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
  • Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
  • U-Net: Convolutional Networks for Biomedical Image Segmentation
  • Attention Is All You Need

• Semi-Supervised Learning

  • MixMatch: A Holistic Approach to Semi-Supervised Learning :: Code
  • FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence
  • AdaMatch: A Unified Approach to Semi-Supervised Learning and Domain Adaptation :: Code

• Generative Models

  • Generative Adversarial Networks (NIPS 2014) :: Code

  • Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks (ICCV 2017)

• Transformers

  • Vanilla Transformer
  • Vision Transformer (ViT)
  • DETR
  • DINO
  • DeiT
  • Swin-Transformer
  • MLP-Mixer

• Self-supervised learning

  • SimCLR
  • BYOL
  • MoCo
  • SimSiam
  • SwaV

• 3D vision

  • Mesh deformation
  • 3D-R2N2
  • GAL
  • Pixel2Mesh

Techniques for modeling	About
Augmentation libraries	Albumentations and more
Hyperparameter search	Optuna and more
AutoML	Pycaret / H2O / lightautoml and more
Augmentation techniques	Mixup, Cutmix and more
Ensemble methods	SWA and more
Class imbalance	Balancedsampler, Oversampling and more
Loss functions	Arcface and more
Visualizations	Tensorboard and more
GPU	GPU
Pytorch Lightening	Pytorch Lightening
Mixed/Half precision training	Mixed/Half precision training
Psudolabeling	Psudolabeling