Ayca Duran
Postdoctoral Scholar, Civil and Environmental Engineering
Bio
Ayca is a postdoctoral researcher working at the intersection of visual computing, artificial intelligence, and environmental design. Her current research focuses on developing novel methodologies for analyzing and editing building‑façade images to improve the environmental sustainability of buildings. She leverages deep‑learning models for vision tasks, embeds geometric information into image‑editing workflows, and fine-tunes diffusion models to enable controlled alterations in architectural scenes.
Ayca holds bachelor’s and master’s degrees in architecture from METU and a doctoral degree from ETH Zurich. During her studies, she was a visiting researcher at Stanford University and an exchange student at the Polytechnic University of Milan. Her graduate research contributed to the Powering the City Module of Future Cities Lab Global and a tool-development project for urban building energy models. Her research interests have been recognized through doctoral and postdoctoral fellowships from the Institute of Technology in Architecture (ITA), Fulbright Program, Swiss National Science Foundation, and Design++. Beyond her research, she has contributed to the teaching of architectural design studios and environmental design courses at both ETH Zurich and METU.
Honors & Awards
-
SNSF Postdoc.Mobility, Swiss National Science Foundation (2026)
-
Institute of Technology in Architecture (ITA) Doctoral Fellowship, ITA, ETH Zurich (2022)
-
Fulbright PhD Grant, Fulbright (2021)
-
German Academic Exchange Service (DAAD) Master’s Scholarship, DAAD (2020)
-
Excellence in Graduation (Highest GPA at bachelor graduation), Department of Architecture, METU (2019)
Professional Education
-
Doctor of Philosophy, ETH Zurich (2026)
All Publications
-
Automated facade solar retrofits using feed-forward 3D reconstruction and generative image editing
AUTOMATION IN CONSTRUCTION
2027; 193
View details for DOI 10.1016/j.autcon.2026.107294
-
Deep learning for BIPV segmentation on facades: Comparison with human annotations across facade designs
BUILDING AND ENVIRONMENT
2026; 293
View details for DOI 10.1016/j.buildenv.2026.114292
View details for Web of Science ID 001688527400001
-
Pretrained Video Models as Differentiable Physics Simulators for Urban Wind Flows
European Conference on Computer Vision (ECCV)
2026
View details for DOI 10.1007/978-3-032-37010-5_12
-
A review on artificial intelligence applications for facades
BUILDING AND ENVIRONMENT
2025; 269
View details for DOI 10.1016/j.buildenv.2024.112310
View details for Web of Science ID 001374262200001
https://orcid.org/0000-0001-6027-2962