ifmanzhang/huangquan-landscapeconcept-dataset
Huangquan LandscapeConcept Dataset This dataset contains the image and caption material used for the Huangquan Road visual concept LoRA workflow. Contents 157 PNG images 157 paired text caption files Original archive: GameLandscapeConcept_Text (1).zip Compressed archive size: approximately 113.28 MB SHA256: 0FF43EBBB2DE0403BB388901D3EB825E90C41D1E41ADC5C0F1FAFD7A4F1415B0 Project Use The dataset was used to train the project's… See the full description on the dataset page: https://huggingface.co/datasets/ifmanzhang/huangquan-landscapeconcept-dataset.
Huangquan LandscapeConcept Dataset
This dataset contains the image and caption material used for the Huangquan Road visual concept LoRA workflow.
Contents
- 157 PNG images
- 157 paired text caption files
- Original archive:
GameLandscapeConcept_Text (1).zip - Compressed archive size: approximately 113.28 MB
- SHA256:
0FF43EBBB2DE0403BB388901D3EB825E90C41D1E41ADC5C0F1FAFD7A4F1415B0
Project Use
The dataset was used to train the project's LandscapeConcept.safetensors LoRA for the underworld landscape style used by Huangquan Road. In the current game, that LoRA is used through the local ComfyUI/FLUX image-generation pipeline.
Dataset Status
This is project visual concept data prepared for assessment review and project documentation. It should be discussed separately from the preliminary Pinterest ink/texture LoRA prototype, which is not part of the current game runtime and is not project-original data.
Related Project Resources
- Main project repository:
https://git.arts.ac.uk/y-zhang0320251/MML_Final.git - Project datasheet: see
DATASHEET.mdin the main repository. - Technical notes: see
docs/TECHNICAL_HIGHLIGHTS.mdin the main repository.