About

Hello, I am Yiyu Liu. I am an incoming Ph.D. student at Shenzhen Loop Area Institution (SLAI) and Southeast University. My doctoral research is dedicated to the forefront of Large Language Models (LLMs), Vision-Language Models (VLMs), Speech Models, and Affective Computing.

Previously, I served as a Research Assistant at Greater Bay Area University, where I primarily explored Video Camouflaged Object Detection (VCOD) and low-level vision. My research perspective is shaped by a highly diverse background: I hold a B.S. in Applied Physics from Southeast University, an M.S. in Computer Vision from Waseda University, and have gained valuable industry experience as a Cyber Security Engineer at China Telecom.

🔥 News

[2026.05] I will be joining Shenzhen Loop Area Institution & SEU as a Ph.D. student!
[2026.04] Submitted our latest work on VCOD to Visual Intelligence 2026.

📝 Selected Publications

[YUV20K: A Complexity-Driven Benchmark and Trajectory-Aware Alignment Model for Video Camouflaged Object Detection] Yi-Yu Liu, et al.
Under Review for Vision Intelligence(VI), 2026.
[PDF] [Code] [Project Page]
[Multi-Level Spatial-Temporal Feature Aggregation and Alignment-Based Selective Residual Dense Propagation Module for HDR Video Reconstruction ] Yi-Yu Liu, et al.
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024.
[PDF]

💻 Projects & Datasets

The YUV20K Dataset Large-scale benchmark for Video Camouflaged Object Detection Led the construction and annotation of YUV20K, a complexity-driven dataset designed to tackle complex motion scenarios in VCOD tasks.

🎓 Education & Experience

Shenzhen Loop Area Institution & Southeast University (2026.09 - Present)
Ph.D. in Artificial Intelligence
Greater Bay Area University (2025 - 2026)
Research Assistant (Computer Vision)
China Telecom (2024.07 - 2025.09)
Network Security Management Engineer
Waseda University M.S. in Computer Vision
Southeast University B.S. in Applied Physics