PrismLayersPro100k

# PrismLayers: Open Data for High-Quality Multi-Layer Transparent Image Generative Models ## PrismLayersPro100k `PrismLayersPro100k` is a high-quality dataset designed for training and evaluating multi-layer transparent image generation models. It contains 100,000 samples, each with multiple layers including foreground objects and background scenes. The dataset aims to facilitate research in transparent object compositing, layer decomposition, and text-aware image generation. Below are several visual statistics that illustrate key characteristics of the dataset: ### 1. Character Count Distribution This plot shows the distribution of the number of characters in text layers across the dataset. It helps understand the typical length and complexity of textual content embedded in images.  --- ### 2. Layer Count Distribution This plot summarizes how many layers (foregrounds, texts, etc.) are present per image in the dataset.  --- ### 3. Text Area Ratio Distribution This plot illustrates the proportion of image area occupied by text regions. It indicates how dominant or subtle the text is in the dataset.  --- ### 4. Text Layer Presence Distribution This figure shows the distribution of images containing varying numbers of text layers. It helps quantify how frequently textual content is included in compositing.  --- These visualizations provide insights into the diversity and complexity of the dataset, making it a valuable resource for generative modeling tasks that require fine-grained control over image structure and composition.