Hi, I'm Hao-Wei Chen. I am a PhD student at National Taiwan University (NTU), supervised by Prof. Chun-Yi Lee. My research focuses on low-level vision tasks and image restoration. I am currently working on robot learning, optimal control, and foundation models for embodied intelligence.
I received both my B.S. and M.S. degrees from National Tsing Hua University, HsinChu, Taiwan in 2019 and 2022, respectively, all in Computer Science.
I'm interested in computer vision, machine learning, and image restoration. Currently exploring robot learning, optimal control, and foundation models for embodied intelligence.
An efficient all-around vision state space model for image restoration that leverages the Mamba architecture to achieve strong performance across multiple restoration tasks.
A novel approach to image super-resolution that exploits holistic semantics and diffusion priors for high-quality image reconstruction.
Exploiting underlying similarities across different image restoration tasks using lightweight adapters for efficient multi-task restoration.
Addressing ambiguity in 6D object pose estimation by formulating the problem as score-based diffusion on the SE(3) manifold.
Boosting flow-based generative super-resolution models by introducing a learned prior to improve the quality and fidelity of generated high-resolution images.
We use self-attention mechanism to enhance the sharpness and clearness of arbitrary-scale super resolution images.
