Hi, I’m preparing a dataset for style‑transfer training (e.g., converting ordinary photos to Van Gogh‑style images) using the expected format:
dataset/
├── images/ # Target images and their captions
│ ├── image_001.jpg
│ ├── image_001.txt
│ ├── image_002.jpg
│ ├── image_002.txt
│ └── ...
└── control/ # Control images
├── image_001.jpg
├── image_002.jpg
└── ...
My questions are as follows:
Control images – Should these be the original photos (the input to the style‑transfer process)?
Target images – Should these be the styled versions (e.g., Van Gogh‑fied) that correspond to each original photo?
Captions – What should the .txt captions describe? Something like “convert this image to Van Gogh style,”?
Could you please confirm whether this is the intended use?
Thank you for your excellent work!
Hi, I’m preparing a dataset for style‑transfer training (e.g., converting ordinary photos to Van Gogh‑style images) using the expected format:
dataset/
├── images/ # Target images and their captions
│ ├── image_001.jpg
│ ├── image_001.txt
│ ├── image_002.jpg
│ ├── image_002.txt
│ └── ...
└── control/ # Control images
├── image_001.jpg
├── image_002.jpg
└── ...
My questions are as follows:
Control images – Should these be the original photos (the input to the style‑transfer process)?
Target images – Should these be the styled versions (e.g., Van Gogh‑fied) that correspond to each original photo?
Captions – What should the .txt captions describe? Something like “convert this image to Van Gogh style,”?
Could you please confirm whether this is the intended use?
Thank you for your excellent work!