AID-AppEAL: Automatic Image Dataset and Algorithm for Content Appeal Enhancement and Assessment Labeling (ECCV 2024)
We crawlled two stock image websites for image thumbtails using image_crawling.py and filter them based on the criteria listed in paper Sec. 3.1 (Fig. 2) using image_filtering.py. For more details, please refer to the code and the paper section.
To generate our synthetic datasets, first install Automatic1111 Stable Diffusion web UI to <sd_webui_path>. Then copy our unappealing/unappealing embeddings located at here under embeddings to <sd_webui_path>/embeddings and copy our local synthetic_appeal_dataset_creation.py to <sd_webui_path>/scripts.