The success of a product in the market depends not only on its quality, but also on how well it aligns with the brand style. Designers face the challenge of creating products that stand out from competitors while maintaining brand recognition.
To address this issue, scientists from the University of Carnegie-Mellon have introduced a groundbreaking technology that can assist designers in their work. They have developed a fully automated neural network called Bignet (Brand Identification Graph Neural Network) that can identify the visual features associated with a brand by analyzing vector images of products[source].
Before the introduction of Bignet, there was no method to automatically extract style features using machine learning. Designers relied on creating rules in their own minds, making it difficult to formulate and pass on those rules to other products.
In order to test Bignet’s capabilities, the research team applied the model to popular Apple and Samsung smartphones and achieved 100% accuracy in distinguishing between the two brands, identifying specific features such as screen clearance and camera placement.
To demonstrate Bignet’s adaptability and generalization across different products and design scales, the team evaluated it on 10 automobile brands. The model accurately identified the style features of Audi, BMW, and Mercedes-Benz, indicating that these manufacturers exhibit a higher degree of style consistency compared to budget car brands.
Bignet’s ability to identify brand design features can greatly save time for designers, as companies will no longer have to rely solely on experienced individuals to understand the brand style[source].
Currently, Bignet works with two-dimensional images only. However, researchers have plans to expand the technology’s capabilities to three-dimensional images and develop a model that can determine not just brand identity, but also other product characteristics, such as distinguishing features of sports cars or SUVs.
The development of Bignet highlights how artificial intelligence can ensure accuracy and consistency in design. By automatically identifying key elements that make a brand recognizable and appealing, this technology offers a significant improvement over traditional subjective assessments and designer experience. It paves the way for a new era in design where machines and human talent collaborate to create innovative and effective products.