Baiang Li

About Me

I am currently an incoming PhD student in Princeton University.
I have the honor to collaborate closely with Jian Wang@ Snap Research and Xiaogang Xu@ CUHK.
My goal for research is to build a deep synergy between AI and our physical world, allowing deep learning methods to illuminate new perspectives in diverse scientific domains. Inspired by the notion that mathematics—while elegant—may only represent a local optimum in our search for fundamental truths, I believe that black-box models, unfettered by strictly human-defined formalism, can reveal deeper patterns and structures. A striking example is how AlphaGo Zero, trained purely through self-play, outperformed its predecessor that relied on human heuristics—suggesting that data-driven discovery can transcend the limits of established knowledge.
However, defining these black-boxes is itself a nontrivial process, and we must balance the power of automated approaches with principled design to avoid mere ‘blind’ exploration. So another key focus of my research is to create domain-specific architectures tailored for physical systems, leveraging both the intrinsic constraints of natural laws and the emergent capabilities of learning-based models. I hope to move beyond local minima in our mathematical and physical understanding, forging pathways toward richer, more universal insights in science and engineering.
--Written on January 20, 2025.
I am open to discussions about potential collaborations, reviews and guidance.

Let's build the PHYSICAL world with AI!

📸Research Interests

Computer Vision: Low-level and High-level computer vision tasks (image detection, segmentation, recognition, etc.)
Image Processing: image/video quality enhancement, image/video editing/manipulation, and generation.
Cross domain between computer vision and computational photography: Building bridges between deep learning models and computational imaging optics.
Deep Learning Driven Fundamental Science: Utilizing black-box/data-driven computational models to transcend traditional mathematical and physical constraints, aiming to uncover deeper patterns and structures that challenge conventional local minima in established theories to reveal emergent phenomena and novel insights in foundational science.

🔥News

[Feb. 2025] Get officially accepted and become a PhD student in Princeton University!
[Jul. 2024] I will visit the Computational Imaging Lab@Princeton University and conduct research work on computational photography as a research assistant.
[Jun. 2024] We release our new work about low dynamic range image enhancement.
[Mar. 2024] One paper about image matting is accepted to Siggraph 2024.
[Feb. 2024] We release our new work about image deraining.
[Nov. 2023] I will start my internship at OpenMMLab@Shanghai AI Lab.
[Aug. 2023] Our paper about low-light stereo image enhancement is accepted to ACM-MM 2023.

🖨️Publications&Preprints

arXiv

Sagiri: Low Dynamic Range Image Enhancement with Generative Diffusion Prior

Baiang Li, Sizhuo Ma, Yanhong Zeng, Xiaogang Xu, Youqing Fang, Zhao Zhang, Jian Wang*, Kai Chen* (*Corresponding authors).

Paper Project page Code

Siggraph

Matting by Generation

Zhixiang Wang, Baiang Li, Jian Wang*, Yu-Lun Liu, Jinwei Gu, Yung-Yu Chuang, Shin'ichi Satoh

Paper Code Project Page Video

arXiv

Diving Deep into Regions: Exploiting Regional Information Transformer for Single Image Deraining

Baiang Li, Zhao Zhang*, Huan Zheng, Xiaogang Xu, Yanyan Wei, Jingyi Zhang, Jicong Fan, Meng Wang

Paper Code

ACM-MM

Dynamic Grouped Interaction Network for Low-Light Stereo Image Enhancement

Baiang Li, Huan Zheng, Zhao Zhang*, Yang Zhao, Zhongqiu Zhao, Haijun Zhang

Paper Code Project Page

🏆Challenge Reports

👓Services

Conference Reviewers

Program Committee

Association for the Advancement of Artificial Intelligence (AAAI) 2025

Journal Reviewers

💻Internships

Computational Imaging Lab@Princeton University, New Jersey, USA

July 2024 – Current

Visiting Student Research Collaborator (VSRC)
Advisor: Felix Heide
Topic: Computational Photography

OpenMMLab @Shanghai AI Lab , Shanghai, China

Nov. 2023 – Mar. 2024

Research Intern in MMagic group
Advisor: Yanhong Zeng
Topic: Image/video restoration and generation

Zhejiang Lab, Hangzhou, China

July 2023 – Sep. 2023

Research Intern
Advisor: Xiaogang Xu and Yulun Zhang
Topic: low-light image/video enhancement and image/video restoration

Key Laboratory of Knowledge Engineering
with Big Data(Ministry of Education), Hefei, China

Aug. 2022 – July 2023

Research Assistant
Advisor: Zhao Zhang
Topic: low-light image/video enhancement and image/video restoration