Saeedeh Barzegar Khalilsaraei

I studied Computer Engineering at the University of Guilan, Rasht, Iran. Then, I pursued a Master's degree in Artificial Intelligence at Kharazmi University, Tehran, Iran. Currently, I am affiliated with the Institute of Visual Computing while pursuing a PhD in a joint degree between Graz University of Technology and Nanyang Technological University. My research focuses on providing an accurate and coarse mathematical representation in the 2D/3D domain to improve the computer-aided design (CAD) process. Leveraging geometric deep learning techniques, I am developing models that are both computationally efficient and easy to manipulate, enabling designers to work more intuitively with complex data.

s.barzegarkhalilsaraei[at]tugraz.at

Selected Publication

1. Saeedeh Barzegar Khalilsaraei and Ursula Augsdörfer. "An Inception-based Variational Autoencoder for Curves Generation and Interpolation." International Conferences in Central Europe on Computer Graphics, Visualization and Computer Vision, 2025. [paper] [code]
2. Alexander Komar, Saeedeh Barzegar Khalilsaraei and Ursula Augsdörfer. "2D B-spline Curve Reconstruction using Convolutional Auto-Encoders and Distance Fields." International Conferences in Central Europe on Computer Graphics, Visualization and Computer Vision, 2025. [paper]
3. Saeedeh Barzegar Khalilsaraei, Alexander Komar, Jianmin Zheng, and Ursula Augsdörfer. "InceptCurves: Curve Reconstruction Using an Inception Network." The Visual Computer, 2024, 1–11. Springer. [paper][code]
4. Alexander Komar, Michael Kammerer, Saeedeh Barzegar Khalilsaraei, and Ursula Augsdörfer. "LatEd: A Geometric Latent Vector Editor." Vácval Skala-UNION Agency, 2024. [paper]
5. Saeedeh Barzegar Khalilsaraei, Manoochehr Kelarestaghi, and Farshad Eshghi. "An Ensemble-Based Method for Vehicle Color Recognition." Open Science Index 16, vol. 23 (2022): 6. [paper]