Chunlin Yu
I'm a third-year master's student at ShanghaiTech University, where I am currently advised by Prof. Jingya Wang.
Before pursuing the master's degree, I obtained my bachelor's degree from ShanghaiTech University.
Email / CV / Google Scholar / Github
My research interests include self-supervised learning and human-centered vision, including but not limited to MLLMs, deep clustering, biometric recognition, 2D/3D human-object interaction, and affordances.
We introduce SeqAfford, a Multi-Modal Language Model (MLLM) capable of serialized affordance inference implied in human instructions.
A challenging benchmark CCGait that captures realistic appearance changes over expanded time and space, as well as a hybrid framework HybridGait.
We propose ConAff Neighborhoods for more context-rich neighbor retrievals as well as a progressive Boundary Filtering strategy for noise-resilient neighborhoods.
Using a biological-inspired network involving a dynamical memory model and an adaptive working model to refresh and consolidate the knowledge on the fly.
A high-order transformer that captures omni-relational information for person re-identification.
Fored from jonbarron's website.