研0卑微 主要方向是机器学习-计算机视觉，求一个入门学习路线，前期应该看哪些入门的论文

Question

研0卑微求问
主要方向是机器学习-计算机视觉，求一个入门学习路线，前期应该看哪些入门的论文？

Answer 1

我把这些年经典的CV论文题目发给你，其中部分可以在B站李沐里面听精讲

• 2010年：Noise-contrastive Estimation: a New Estimation Principle for Unnormalized Statistical Models
• 2012年：ImageNet Classification with Deep Convolutional Neural Networks
• 2013年：Visualizing and Understanding Convolutional Networks
• 2015年
  • Very Deep Convolutional Networks for Large-Scale Image Recognition
  • Going Deeper with Convolutions
  • FaceNet: a Unified Embedding for Face Recognition and Clustering
• 2016年
  • Rethinking the Inception Architecture for Computer Vision
  • Deep Residual Learning for Image Recognition
  • Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• 2017年：Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
• 2018年
  • From Recognition to Cognition: Visual Commonsense Reasoning
  • Focal Loss for Dense Object Detection
  • Relational Inductive Biases, Deep Learning, and Graph Networks
• 2019年
  • Objects As Points
  • RandAugment: Practical Automated Data Augmentation with a Reduced Search Space
  • Semantic Image Synthesis with Spatially-Adaptive Normalization
• 2020年
  • Denoising Diffusion Probabilistic Models
  • Designing Network Design Spaces
  • An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
  • Training Data-efficient Image Transformers & Distillation Through Attention
  • NeRF: Representing Scenes As Neural Radiance Fields for View Synthesis
  • Bootstrap Your Own Latent: a New Approach to Self-supervised Learning
  • A Simple Framework for Contrastive Learning of Visual Representations
  • Conditional Negative Sampling for Contrastive Learning of Visual Representations
  • Momentum Contrast for Unsupervised Visual Representation Learning
  • Generative Pretraining from Pixels
• 2021年
  • Do Vision Transformers See Like Convolutional Neural Networks?
  • BEiT: BERT Pre-Training of Image Transformers
  • Swin Transformer: Hierarchical Vision Transformer Using Shifted Windows
  • RepVGG: Making VGG-style ConvNets Great Again
  • An Empirical Study of Training Self-Supervised Vision Transformers
  • Diffusion Models Beat GANs on Image Synthesis
• 2022年
  • A ConvNet for the 2020s
  • Natural Language Descriptions of Deep Visual Features
  • Vision Models are More Robust and Fair When Pretrained on Uncurated Images Without Supervision
  • Block-NeRF: Scalable Large Scene Neural View Synthesis
  • VICReg: Variance-Invariance-Covariance Regularization for Self-Supervised Learning
  • Masked Autoencoders are Scalable Vision Learners
  • The Effects of Regularization and Data Augmentation are Class Dependent
  • Instant Neural Graphics Primitives with a Multiresolution Hash Encoding
  • Pix2seq: a Language Modeling Framework for Object Detection
  • An Improved One Millisecond Mobile Backbone
  • Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding
  • Swin Transformer V2: Scaling up Capacity and Resolution
  • Scaling Autoregressive Models for Content-Rich Text-to-Image Generation