Paper Reading Notes

• Deep Learning
  • Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics
• Object Detection
  • Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation
  • Fast R-CNN
  • Bridging the Gap Between Anchor-Based and Anchor-Free Detection via Adaptive Training Sample Selection (ATSS)
  • Probabilistic Anchor Assignment with IoU Prediction for Object Detection
  • Focal Loss for Dense Object Detection (RetinaNet)
  • End-to-End Object Detection with Transformers (DETR)
  • Mask R-CNN
• Monocular 3D Object Detection
  • Rethinking Pseudo-LiDAR Representation
• Generative Models
  • Auto-Encoding Variational Bayes
  • InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
  • Improved Techniques for Training GANs
  • Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks
  • Image-to-Image Translation with Conditional Adversarial Networks (Pix2Pix)
  • High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs (Pix2PixHD)
  • Image Based Virtual Try-On Network from Unpaired Data
• Compression (Image)
  • Reducing the Dimensionality of Data with Neural Networks
  • End-To-End Optimized Image Compression
  • Soft-to-Hard Vector Quantization for End-to-End Learning Compressible Representations
  • Conditional Probability Models for Deep Image Compression
  • Generative Adversarial Networks for Extreme Learned Image Compression
  • Causal Contextual Prediction for Learned Image Compression
• Compression (Video)
  • An End-to-End Learning Framework for Video Compression
  • Improving Deep Video Compression by Resolution-Adaptive Flow Coding
• Transformer
  • Attention is All You Need
  • An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale (ViT)
  • Pre-Trained Image Processing Transformer
  • A Survey on Visual Transformer
  • Learning Texture Transformer Network for Image Super-Resolution
  • Taming Transformers for High-Resolution Image Synthesis
  • Temporal-Channel Transformer for 3D Lidar-Based Video Object Detection in Autonomous Driving
  • Residual Non-Local Attention Networks for Image Restoration
  • CPTR: Full Transformer Network for Image Captioning
  • Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet
• Super Resolution
  • MuCAN: Multi-Correspondence Aggregation Network for Video Super-Resolution
• Other Papers
  • PifPaf: Composite Fields for Human Pose estimation
  • Non-Local Neural Networks
  • RAFT: Recurrent All-Pairs Field Transforms for Optical Flow