A5013352663- Design Education and Practice
- Creativity in Education and Neuroscience
- Aesthetic Perception and Analysis
- Innovative Human-Technology Interaction
- Color perception and design
- Manufacturing Process and Optimization
- Architecture and Computational Design
- Ethics and Social Impacts of AI
- Web Data Mining and Analysis
- Big Data and Business Intelligence
- Plant and Biological Electrophysiology Studies
- Topic Modeling
- 3D Shape Modeling and Analysis
- Augmented Reality Applications
- Innovation, Sustainability, Human-Machine Systems
- Software Engineering Research
- 3D Surveying and Cultural Heritage
Toyota Industries (United States)
2023-2025
Toyota Research Institute
2023-2025
Designers often struggle to sufficiently explore large design spaces, which can lead fixation and suboptimal outcomes. Here we introduce DesignAID, a generative AI tool that supports broader space exploration by first using language models produce range of diverse ideas expressed in words, then image generation software create images from these words. This innovative combination AI-based capabilities allows human-computer pairs rapidly set visual concepts without time-consuming drawing. In...
Nature often inspires solutions for complex engineering problems, but it is challenging designers to discover relevant analogies and synthesize from them. Here, we present an end-to-end system, BioSpark, that generates biological-analogical mechanisms provides interactive interface comprehension ideation. From a small seed set of expert-curated mechanisms, BioSpark's pipeline iteratively expands them by constructing traversing organism taxonomies, aiming overcome both data sparsity in expert...
With recent advancements in the capabilities of Text-to-Image (T2I) AI models, product designers have begun experimenting with them their work. However, T2I models struggle to interpret abstract language and current user experience tools can induce design fixation rather than a more iterative, exploratory process. To address these challenges, we developed Inkspire, sketch-driven tool that supports prototyping concepts analogical inspirations complete sketch-to-design-to-sketch feedback loop....
Through iterative, cross-disciplinary discussions, we define and propose next-steps for Human-centered Generative AI (HGAI). We contribute a comprehensive research agenda that lays out future directions of spanning three levels: aligning with human values; assimilating intents; augmenting abilities. By identifying these next-steps, intend to draw interdisciplinary teams pursue coherent set emergent ideas in HGAI, focusing on their interested topics while maintaining big picture the work landscape.
Generative AI (GenAI) models excel in their ability to recognize patterns existing data and generate new unexpected content. Recent advances have motivated applications of GenAI tools (e.g., Stable Diffusion, ChatGPT) professional practice across industries, including product design. While these generative capabilities may seem enticing on the surface, certain barriers limit practical application for real-world use industry settings. In this position paper, we articulate situate within two...
Consumers' emotional and cognitive attachment to product design plays a pivotal role in influencing purchasing choices. Therefore, designers incorporate this signal as they develop new products. The goal of our work is reduce the psychological distance between consumers automotive concept process. While generative AI models hold potential amplify creativity, these do not have any specialized knowledge. In work, we developed novel framework system that combines machine learning, human...
Labeling short, unstructured texts is generally performed by sequentially identifying codes and assigning them to segments of text based on viewing a small sample data. In this greedy approach, coders risk overlooking important code ideas must perform the tedious task iteratively revising initial set, sometimes response assignments, as new themes emerge. To address this, we propose CodeML, machine learning-assisted (ML) coding interface that identifies multiple in response, which are...
Nature is often used to inspire solutions for complex engineering problems, but achieving its full potential challenging due difficulties in discovering relevant analogies and synthesizing from them. Here, we present an end-to-end system, BioSpark, that generates biological-analogical mechanisms provides interactive interface comprehend synthesize BioSpark pipeline starts with a small seed set of expands it using iteratively constructed taxonomic hierarchies, overcoming data sparsity manual...