| Title |
An Analysis on Consistency of Brand Identity with AI-based Image Classification |
| Authors |
한유진(Han, Yoojin) ; 이현수(Lee, Hyunsoo) |
| DOI |
https://doi.org/10.14774/JKIID.2019.28.6.138 |
| Keywords |
Brand Identity; Retail Design; Deep Learning; Image Classification |
| Abstract |
The aim of this study is to investigate a way of determining consistency of brand identity in store design with deep learning-based technology. The concept of brand identity has been aroused in order to set brands apart from tough competitors in global marketplace. Especially, brand space has been highlighted as a strong communicator of consistent brand identity. However, in terms of spatial environments, brand identity has been studied adopting qualitative approaches. This paper look into a way of quantifying consistency of brand identity adopting deep learning-based image classification model. Following the case of Starbucks and Bluebottle ? two brands creating constant and strong brand identity attached to worldwide brand spaces ? auto image classification was conducted with TensorFlow to investigate brand identity congruence. Departing from training image recognition model, 3 tests were conducted to demonstrate the train model. This paper demonstrates that a consistent, coherent, and strong brand identity attached to store design can be trained and recognized with deep learning-based technology. This research also suggests the wider usage of this model in branding and interior design. |