http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
적대적 생성 모델을 활용한 AI 기반 도시 블록 생성에 관한 연구-서울 사례를 중심으로-
박철웅(Park, Chul Woong),오무열(Oh, Mooyeol),이소정(Lee, Sojeong),변기영(Byun, Giyoung),김영철(Kim, Youngchul) 한국생태환경건축학회 2021 한국생태환경건축학회 학술발표대회 논문집 Vol.21 No.2
Urban block design demands complexly examining legal and aesthetic standards including configuration of roads, land use and parcels. Because of various constraints, urban block design requires time-consuming process of few experts. This study suggests AI-driven model for effective urban block design. We extract image files that contain information of land use, road and parcel configuration in urban block as RGB to train Deep Convolutional Generative Adversarial Network (DCGAN). In conclusion, the trained DCGAN model generates urban block image based on configuration of road, land use and block shape in Seoul. Although further studies are necessary, this study suggests that DCGAN is a possible approach to develop applicable AI-driven urban design.