Dr. Guo Cheng

Assistant Professor
Department of Computer Science
Texas A&M University–Corpus Christi  ·  Starting August 2026

Ph.D. in Computer Science, Purdue University, 2022
Previously: Machine Learning / Software Engineer at Bloomberg L.P., New York

Research Interests: Computer Vision · Deep Learning · Artificial Intelligence · Autonomous Driving

I am hiring Ph.D. students starting Fall 2026 / Spring 2027 / Fall 2027. If you are interested in computer vision, AI/ML, autonomous driving or robotics, please read the prospective students page and reach out.

About me

I am an incoming tenure-track Assistant Professor of Computer Science at Texas A&M University–Corpus Christi (TAMU-CC), starting in August 2026. My research is in computer vision, deep learning, and autonomous driving, with a particular focus on perception and scene understanding for safety-critical systems.

I received my Ph.D. in Computer Science from Purdue University in 2022, advised by Professor Jiangyu Zheng, with a dissertation titled "Sequential Semantic Segmentation of Streaming Scenes for Autonomous Driving." My work has been published in IEEE T-ITS, IEEE T-IV, ICPR, IV, and ITSC. I earned an M.S. in Statistics from Stony Brook University and a B.S. in Information and Computer Science from Wuhan Textile University. Prior to joining TAMU-CC, I was a Software Engineer at Bloomberg L.P. in New York.

At TAMU-CC, I am building a research group at the intersection of computer vision and autonomous systems. I am actively recruiting Ph.D. students. Please see the prospective students page.

Research interests

News

Selected projects

Sequential Semantic Segmentation for Streaming Driving Scenes

Ph.D. dissertation work on memory-efficient, sequential semantic segmentation of high-resolution driving video, with applications to path and speed planning for autonomous vehicles. Published in IEEE T-ITS and ICPR.

PyTorch · TensorFlow · OpenCV · CNN/Transformer

Vision-Language Models under Perceptual Degradation

Recent preprint studying when and how VLMs lose semantic alignment under noise, blur, weather, and other perceptual perturbations — with implications for safety-critical perception pipelines.

VLMs · multimodal · robustness

Big-Video Mining of Driving Appearances

Mining large-scale driving footage to model weather and illumination, supporting all-weather road-edge detection and robust perception. Published in T-IV and ITSC.

video mining · sensor fusion · ADAS

Teaching

Services

Conference reviewer: ICPR (2020–2022), IEEE IV (2018–2022), IEEE ITSC (2018–2022).

Journal reviewer: The Journal of Supercomputing (2026), The Visual Computer (2025), Multimedia Tools and Applications (2025), IEEE TPAMI (2021), IEEE T-ITS (2020–2021), IEEE T-IV (2019–2021), IEEE TVT (2019).