---
# System prepended metadata

title: 'Weather GAN: Multi-Domain Weather Translation Using Generative Adversarial Networks'
tags: [Reid paper survey]

---

# Weather GAN: Multi-Domain Weather Translation Using Generative Adversarial Networks
###### tags: `Reid paper survey`

- unknow
    - minimax game [11].
    - model collapse
        - in unsupervised training, the generator may map 
          most data of source domain to a few samples of 
          target domain, which makes the generated results 
          lack of diversity. This phenomenon is also called 
          model collapse.
    - how to know attention
    - weakly supervion
    - Dataset set 是其他新的?
    - D 在最後比的是label 還是天氣種類
        - 天氣種類
# Model Architecture
- the weather translation task is cast as an unsupervised image translation task
- three equally important tasks: 
    - (1) identifying objects of the image. 
    - (2) locating the region of interest from these 
      recognized objects. 
    - (3) the targeted region of interest is translated 
      correctly