Creative Blends of Visual Concepts

2025-02-22

AI Summary

This work addresses the challenge of generating coherent visual fusion images from abstract concepts—specifically, how to seamlessly integrate two heterogeneous visual elements to convey deep metaphorical meaning. We propose the first AI-assisted design framework that jointly integrates metaphorical reasoning, commonsense-knowledge-guided intent alignment, and generative visual fusion. Our method leverages commonsense knowledge bases (e.g., ConceptNet) and large language models to calibrate user design intent, then employs attribute-overlap modeling to drive Stable Diffusion for semantically consistent single- or multi-object visual fusion. Evaluation shows a 37% improvement in metaphorical expressiveness and enhanced creative output quality. A user study (N=24) confirms reduced cognitive load and improved design interpretability and controllability. The core contribution is the first systematic unification of computational metaphor modeling, commonsense-aligned intent grounding, and diffusion-based generation—establishing a novel paradigm for abstract concept visualization.

Technology Category

• Computer Vision: Visual Reasoning & Symbolic Representations
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Knowledge Representation and Reasoning: Common-Sense Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Visual blends combine elements from two distinct visual concepts into a single, integrated image, with the goal of conveying ideas through imaginative and often thought-provoking visuals. Communicating abstract concepts through visual blends poses a series of conceptual and technical challenges. To address these challenges, we introduce Creative Blends, an AI-assisted design system that leverages metaphors to visually symbolize abstract concepts by blending disparate objects. Our method harnesses commonsense knowledge bases and large language models to align designers' conceptual intent with expressive concrete objects. Additionally, we employ generative text-to-image techniques to blend visual elements through their overlapping attributes. A user study (N=24) demonstrated that our approach reduces participants' cognitive load, fosters creativity, and enhances the metaphorical richness of visual blend ideation. We explore the potential of our method to expand visual blends to include multiple object blending and discuss the insights gained from designing with generative AI.

Problem

• AI-assisted visual concept blending
• Reducing cognitive load in design
• Enhancing creativity with metaphors

Innovation

• AI-assisted design system
• Generative text-to-image techniques
• Commonsense knowledge bases