A Creative Industry Image Generation Dataset Based on Captions

TL;DR

This paper introduces a novel dataset for creative industry image generation that includes prompts and sketches across key domains, enabling better controllability and evaluation of generative models.

Contribution

It presents the first dataset covering major creative industry areas with prompt and sketch labels, along with multiple references and fine-grained scoring for improved measurement.

Findings

State-of-the-art models reveal prompt importance over sketches.

Dataset facilitates controllable image generation in creative fields.

Shortcomings of current models are identified.

Abstract

Most image generation methods are difficult to precisely control the properties of the generated images, such as structure, scale, shape, etc., which limits its large-scale application in creative industries such as conceptual design and graphic design, and so on. Using the prompt and the sketch is a practical solution for controllability. Existing datasets lack either prompt or sketch and are not designed for the creative industry. Here is the main contribution of our work. a) This is the first dataset that covers the 4 most important areas of creative industry domains and is labeled with prompt and sketch. b) We provide multiple reference images in the test set and fine-grained scores for each reference which are useful for measurement. c) We apply two state-of-the-art models to our dataset and then find some shortcomings, such as the prompt is more highly valued than the sketch.