[游戏开发]Paper Reading-list

Denoising?

• Self-Supervised Image Denoising via Iterative Data Refinement? CVPR22 (unsupervised )
  paper: https://arxiv.org/abs/2111.14358
  code: https://github.com/zhangyi-3/IDR

Super Resolution

• DCLS Deep Constrained Least Squares for Blind Image Super-resolution (physical formulation, optimization,?CVPR22 )
  paper: https://arxiv.org/pdf/2202.07508.pdf
  code: https://github.com/Algolzw/DCLS

• Learning the Degradation Distribution for Blind Image Super-Resolution (GAN, data augmentation； real-world data CVPR22 )
  paper: https://arxiv.org/pdf/2203.04962.pdf
  code: https://github.com/greatlog/UnpairedSR

• Details or Artifacts_ A Locally Discriminative Learning Approach to realistic SR? (GAN; learning? artifact mask; variance CVPR22 )

• ICCV_2021_Fourier_Space_Losses_for_Efficient_Perceptual_Image_Super-Resolution_ICCV_2021_paper? (Frequency loss: amplitude and phase)

Transformer

• U-former
• Restormer

Inpainting

• Bridging Global Context Interactions for High-Fidelity Image Completion?CVPR22? (Transformer, non-overlapping, corase-to-fine)
  paper: https://arxiv.org/abs/2104.00845
  code: https://github.com/lyndonzheng/TFill

Defocus deblur

Shadow?

• benchmarking shadow removal for facial landmark detection and beyond (arxiv, low level for high level,? adversarial attack: data generation)

Sgmentation

• Semi-supervised semantic segmentation with cross-consistency training? CVPR
• GuidedMix-Net：?Semi-supervised semantic segmentation by using labeled images as reference? AAAI22 (labeled-unlabeled image pair interpolation; pseudo mask generation)
• Fourier Domain adaptation for semantic segmentation? CVPR20 (风格转换存在在低频中，进行交换）

Translation?

• ICCV_2021_Frequency_Domain_Image_Translation_More_Photo-Realistic_Better_Identity-Preserving_ICCV_2021_paper? (利用高频的约束保持identity structure； clear issue: over-adapt； 低频： color + illumination // 高频 ： object structure)

Basic Network?

• Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution? (save computation)?
• NeurIPS-2020-fast-fourier-convolution-Paper? (low/high frequency interactively processing)

Knowledge Distillation

• Focal and global knowledge Distillation for detectors (CVPR22, object detection不易蒸馏； Focal : focus on the teacher's critical pixels and channels)

Recognition

• Adaptive focus for efficient video?recognition? (ICCV21 )?知乎?希望从空间的维度上去除冗余，使得网络利用空间维度上更加重要的信息进行动作识别，之前都是从temporal 维度上，这篇也有plus版本， 利用强化学习进行crop