공부해야 될 모델 리스트

 

▶ NLP

• (LSTM)
• transformer
• BERT
• GPT

 

 

▶ Vision AI 

Visual Trnasformer (ViT)

• ViT : paper, 블로그
• MAE (masked auto encoder) pre-trained ViT : 블로그
• Image GPT

 

Generative model

• Diffusion
• GAN

 

Foundation model

• segment anything (잘 정리된 블로그)
  - task : prompt 기반 모델 
  - model : image encoder(MAE pre-trained ViT), prompt encoder(sparse, dense), mask decoder
  - data : 데이터 엔진.
  

 

이외에..

• Object Detection
  - R-CNN, Fast R-CNN, Faster R-CNN, YOLO, SSD 
• Segmentation
  - FCN(fully convolutional network), U-Net, SegNet, DeepLab
• Image Captioning
  - LSTM + CNN
  - NIC (Neural Image Captioning), Show and Tell, Up-Down