Paper_Review_and_Practice

21.01.30~

---

Image

Basic

• Deep Residual Learning for Image Recognition (CVPR 2016) (paper)
  • ResNet / code
  • ResNet_Framework / code
• Batch normalization: Accelerating deep network training by reducing internal covariate shift (PMLR 2015) (paper)
  • 기본적으로 대부분의 코드에 적용됨.

Segmentation & Convolution

• U-Net: Convolutional Networks for Biomedical Image Segmentation (paper)
  • UNet / code
  • UNet_reggression / code

Instace Segmentation

• YOLACT: Real-time Instance Segmentation (paper)
  • YOLACT / official code / 발표자료
    • R-CNN / Fast R-CNN / Faster R-CNN / Mask R-CNN
  • YOLACT++: Better Real-time Instance Segmentation (paper)
  • YolactEdge: Real-time Instance Segmentation on the Edge (Jetson AGX Xavier: 30 FPS, RTX 2080 Ti: 170 FPS) (paper)

GAN

• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks (ICLR 2016) (paper)
  • DCGAN / code
  • DCGAN_Framework / code
• Image-to-Image Translation with Conditional Adversarial Nets (CVPR 2017) (paper)
  • Pix2Pix / code
    • Pix2PixHD : pix2pixHD: High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs (CVPR 2018) (paper)
• SPADE: Semantic Image Synthesis With Spatially-Adaptive Normalization (CVPR 2019) (paper)
• (CVPR 2018) StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation (paper)
• (CVPR 2020) StarGAN v2: Diverse Image Synthesis for Multiple Domains (paper)

Style Transfer

• Image Style Transfer Using Convolutional Neural Networks(Gatys) (paper)
• Perceptual Losses for Real-Time Style Transfer and Super-Resolution(Johnson) (paper)
• Instance Normalization: The Missing Ingredient for Fast Stylization (paper)
• A Learned Representation For Artistic Style (paper)
• Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization(AdaIN) (paper)

Semi-Supervised Learning

• MixMatch: A Holistic Approach to Semi-Supervised Learning (paper)
• ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring (paper)
• FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence (paper)

---

Video

GAN

• Adversarial Video Generation on Complex Datasets (ICLR 2020) (paper)
  • DVD-GAN / 발표자료
    • VGAN / TGAN / MoCoGAN / TGANv2

---

For later

• Clément Godard, Oisin Mac Aodha, Michael Firman, and Gabriel J. Brostow. 2019. Dig-ging into Self-Supervised Monocular Depth Prediction. InInternational Conferenceon Computer Vision (ICCV)