datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
interactive-stage2-30k-v1InteractiveRadiologyTutor_datssetsarticulated-objects-interactive-image-gen
Real-World Evaluation Images for Articulated Objects Interaction Generation
This dataset contains the real-world images used in evaluating DragAPart, a conditional image generator that models interaction with articulated objects.
📦 How to Use It?
Each sample consists of:
original_image_XXX.png: The base image showing an articulated object.
arrow_locations_XXX.npy: A NumPy file containing the arrow coordinates for interaction.
The .npy file stores one arrow as:
[x0, y0… See the full description on the dataset page: https://huggingface.co/datasets/rayli/articulated-objects-interactive-image-gen.interactive_ncaunreal-interactive-devInteractive_Image_Mosaic_Generator
Tile Images Directory
Place small images in this directory to use as mosaic tiles when using the "image_tiles" tile type.
Supported formats:
JPG/JPEG
PNG
BMP
TIFF
Recommendations:
Use small images (preferably square)
Images will be automatically resized to 32x32 pixels
Mix different types of images for varied effects (nature, textures, icons, etc.)
More images = more variety in the mosaic
Examples of good tile images:
Flowers, leaves, stones… See the full description on the dataset page: https://huggingface.co/datasets/nyakura/Interactive_Image_Mosaic_Generator.