https://github.com/ali-vilab/In-Context-LoRA

Please ensure the total image resolution does not exceed 1536 x 1536 pixels. The width and height should be multiples of 64.

For image segmentation, use terms like [SCENE-1], [SCENE-2], [SCENE-3], or [IMAGE1], [IMAGE2], [IMAGE3], etc.

For names, please separate them using the format <name1>, <name2>, <name3>.