Tingyu Lin

Tingyu Lin is a predoctoral researcher at the Computer Vision Lab, Institute of Visual Computing & Human-Centered Technology, Faculty of Informatics, Technische Universität Wien, Austria. She is affiliated with the FWF doctoral program "Visual Analytics and Computer Vision Meet Cultural Heritage (VaCoViCu)" supervised by Prof. Robert Sablatnig, focusing on preserving cultural heritage through automated analysis and visualization of historical photographs and amateur films.

Her research interests include Computer Vision, Deep Learning, and Image Processing, with a strong emphasis on applications in Cultural Heritage Preservation. Recent work spans Historical Image Retrieval, Spectral Reconstruction, and Code Understanding benchmarks.

The 2024 article CodeScope introduces a multidimensional benchmark for LLMs in code analysis, while her 2023 paper HSGAN explores hyperspectral reconstruction using GANs. Earlier works focus on legacy photo editing and spectral analysis from RGB images.

Her research integrates techniques like Generative Adversarial Networks, Multitask Learning, and Feature Composition, contributing to cross-domain applications in cultural heritage and computational linguistics.