http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
AI를 활용한 한국 건축의 발전 가능성 연구 - StyleGAN을 활용한 뉴 타이폴로지 제안 -
천수경(Cheon, Su-Kyeong),김승묵(Kim, Seung-Mook),이상윤(Lee, Sang-Yun) 대한건축학회 2022 대한건축학회 학술발표대회 논문집 Vol.42 No.1
This study proposes a new method of reinterpreting traditional Korean architecture by applying their characteristics to modern architecture. Deviating from the existing studies where subjective opinions dominates, we have implemented non-biased AI to suggest a new way of research. We have analysed images through “GAN” that has been developed to derive general attributes from input images and traverse features across different images without involving our biases. While using GAN, we intended to focus on classifying the images of traditional Korean architectures under designated categories to extract dominant characteristics and to find a possibility of hybridising them with modern architecture. Via this process, we propose the potential of traditional Korean architecture and their development into a new typology.