Chang Liu

A photo of Chang

Hi! I’m Chang, a thesis-based Master’s student specializing in computer vision within the Systems Design Engineering department at the University of Waterloo. I’m co-supervised by Prof. Sirisha Rambhatla and Prof. Alexander Wong as part of the Vision and Image Processing (VIP) Lab and Critical ML Lab.

My ultimate vision is to develop safe and highly capable models that effectively combine multiple modalities, including vision, language, audio, and more. My current research focuses on unsupervised image segmentation, particularly under distribution shifts, with applications in manufacturing and self-driving. This work aligns with my broader vision as I incorporate language as an additional modality to tackle the distributional shift challenge.

Prior to my Master’s, I completed my undergraduate degree at the University of Waterloo, where I collaborated with Prof. Stephen Watt on classifying students’ learning habits and Prof. Sirisha Rambhatla on predicting survival probabilities for patients with alcoholic liver disease.

Email | Twitter/X | LinkedIn

Publications

Chang Liu, Saad Hossain, C Thomas, Henry Lai, Ravi Vemulapalli, Sirisha Rambhatla, and Alexander Wong. LangDA: Adapting visual features with instruction tuning for semantic segmentation. Adaptive Foundation Models @ NeurIPS 2024. [Poster]
Chang Liu, Karim Habashy, Yuchen Pan, Sirisha Rambhatla, and Alexander Wong. Scaled Skip Connections for Semantic Detail Control in Diffusion Models. Conference on Vision and Intelligent Systems (CVIS), December 2024. [Poster]
Chang Liu, Saad Hossain, C Thomas, Henry Lai, Ravi Vemulapalli, Sirisha Rambhatla, and Alexander Wong. Language-guided Domain Adaptive Semantic Segmentation. Conference on Vision and Intelligent Systems (CVIS), Best Vision Paper, December 2024. [Slides]