Computer Vision · Bi-level Optimization

Zhu Liu 刘铸

I am Zhu Liu, a fourth-year Ph.D. student in the School of Software at Dalian University of Technology (DLUT), specializing in Software Engineering under the supervision of Professor Risheng Liu. My research focuses on bi-level optimization, multi-spectral image processing and fusion.

Bi-level optimization serves as an efficient tool for modeling complex coupled learning problems. Concurrently, multimodal image processing and fusion offer an efficient pathway for achieving intelligent perception in unmanned systems, leveraging diverse sensor information to enhance perceptual capabilities. Against this backdrop, my research aims to design efficient bi-level optimization algorithms tailored for vision tasks. The goal is to achieve panoramic understanding and precise perception of real-world scenarios.

Before that, I received the B.E. and M.S.E degrees in software engineering at DLUT, China, in 2019 and 2022.

Email CV Google Scholar
Portrait of Zhu Liu

Dalian University of Technology

01

News

Selected updates

  • 2026Our wide-parallax Gaussian Splatting work was accepted by SIGGRAPH Asia.
  • 2026Our optimization-driven homography estimation work was accepted by IEEE TPAMI.
  • 2026One paper on thermal infrared image super-resolution was accepted by ACM MM 2026 as an Oral paper.
  • 05/2026One paper on point-supervised infrared small target detection was accepted by ICML 2026 (Spotlight Top 2.2%).
  • 03/2026Two papers were accepted by CVPR 2026.
  • 09/2025One paper on bilevel optimization for adversarial learning was accepted by NeurIPS 2025.
  • 02/2025One paper on infrared imaging was accepted by CVPR 2025.
  • 12/2024I was awarded the 2024 CSIG Outstanding Master's Thesis Incentive Program (Top Ten Recipients).
  • 05/2024One paper on bi-level optimization was accepted by ICML 2024.
  • 03/2024One paper on multi-modality image fusion was accepted by IEEE TPAMI 2024.

02

Selected Publications

* Equal contribution

PN-GS wide-parallax Gaussian splatting method overview and reconstruction results

PN-GS: Taming Wide-Parallax Gaussian Splatting via Bi-Level Pose-Gaussian Nested Optimization

Chongyang Feng, Zhu Liu, Xiaofeng Liu, Yuxin Zhang, Xin Fan, Zhongxuan Luo

Proceedings of the SIGGRAPH Asia 2026 Conference, 2026     Paper | Code

Performance comparisons for physics-informed thermal infrared image super-resolution across degradations, perception tasks, and real cameras

Decoupling Corruption from Observation: A Physics-Informed Generative Model for Infrared Image Super-Resolution

Benzhuang Zhang, Zhu Liu, Siyuan Ding, Wengeng Chen, Xingyuan Li, Jinyuan Liu, Long Ma, Risheng Liu

Proceedings of the 34th ACM International Conference on Multimedia, 2026  (Oral)   Paper | Code

Overview of point-supervised infrared small target detection with pseudo-label diffusion
Diffuse to Detect: Bi-Level Sample Rebalancing with Pseudo-Label Diffusion for Point-Supervised Infrared Small Target Detection

Zhu Liu, Yuanhang Yao, Ping Qian, Zihang Chen, Risheng Liu

International Conference on Machine Learning, 2026  (Spotlight, Top 2.2%)

Qualitative infrared small target predictions and semantic feature analysis
Learning with Semantic Priors: Stabilizing Point-Supervised Infrared Small Target Detection

Yuanhang Yao, Ping Qian, Zhu Liu, Long Ma, Weimin Wang

Proceedings of the International Joint Conference on Artificial Intelligence, 2026

Infrared enhancement results across hierarchical degradation scenarios
HiDRA: Hierarchical Degradation Representation and Adaptation with Generative Priors for Enhancing Infrared Vision

Zihang Chen*, Zhu Liu*, Changbo Yan, Haoyuan Xu, Jinyuan Liu, Risheng Liu

Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2026

Dual-space architecture for misaligned multi-modality image fusion
A Hybrid Space Model for Misaligned Multi-modality Image Fusion

Yi Xiao*, Jia Wang*, Zhu Liu, Di Wang, Jinyuan Liu, Risheng Liu

Annual AAAI Conference on Artificial Intelligence, 2026   Paper

EAGLE appearance and geometry-aware visual query localization framework
EAGLE: Episodic Appearance- and Geometry-aware Memory for Unified 2D-3D Visual Query Localization in Egocentric Vision

Yifei Cao, Yu Liu, Guolong Wang, Zhu Liu, Kai Wang, Xianjie Zhang, Jizhe Yu, Xun Tu

Annual AAAI Conference on Artificial Intelligence, 2026   Paper

Progressive prompt fusion framework for infrared vision enhancement
Enhancing Infrared Vision: Progressive Prompt Fusion Network and Benchmark

Jinyuan Liu, Zihang Chen, Zhu Liu, Zhiying Jiang, Long Ma, Xin Fan, Risheng Liu

Advances in Neural Information Processing Systems, 2025   Paper | Code

Optimization curves for bilevel adversarial learning
Bilevel Optimization for Adversarial Learning Problems: Sharpness, Generation, and Beyond

Risheng Liu, Zhu Liu, Weihao Mao, Wei Yao, Jin Zhang

Advances in Neural Information Processing Systems, 2025   Paper

Automatic search space for multi-modality image fusion objectives
Where Elegance Meets Precision: Towards a Compact, Automatic, and Flexible Framework for Multi-modality Image Fusion and Applications

Jinyuan Liu, Guanyao Wu, Zhu Liu, Long Ma, Risheng Liu, Xin Fan

Proceedings of the Thirty-Third International Joint Conference on Artificial Intelligence, 2024   Paper | Code

Data-efficient adversarial learning framework for infrared imaging
DEAL: Data-Efficient Adversarial Learning for High-Quality Infrared Imaging

Zhu Liu, Zijun Wang, Jinyuan Liu , Fanqi Meng, Long Ma, Risheng Liu

IEEE Conference on Computer Vision and Pattern Recognition, 2025   Paper | Code

Overview of infrared-visible image fusion research and applications
Infrared and Visible Image Fusion: From Data Compatibility to Task Adaption

Jinyuan Liu, Guanyao Wu, Zhu Liu, Wang, Jiang, Ma, Zhong, Fan, Risheng Liu

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024   Paper | Code

Convergence comparison for nonconvex bilevel optimization methods
Moreau Envelope for Nonconvex Bi-Level Optimization: A Single-loop and Hessian-free Solution Strategy

Risheng Liu, Zhu Liu, Wei Yao, Shangzhi Zeng, Jin Zhang

International Conference on Machine Learning, 2024   Paper | Code

Task-guided meta-initialized architecture for image fusion
A task-guided, implicitly-searched and metainitialized deep model for image fusion

Risheng Liu, Zhu Liu, Jinyuan Liu, Xin Fan, Zhongxuan Luo

IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024   Paper | Code

Multi-exposure image fusion comparisons and architecture search
Searching a Compact Architecture for Robust Multi-Exposure Image Fusion

Zhu Liu,Jinyuan Liu, Guanyao Wu, Zihang Chen, Xin Fan, Risheng Liu

IEEE Transactions on Circuits and Systems for Video Technology, 2024   Paper | Code

Bi-level adversarial robustness framework for infrared small target detection
Enhancing infrared small target detection robustness with bi-level adversarial framework

Zhu Liu, Zihang Chen, Jinyuan Liu, Long Ma, Xin Fan, Risheng Liu

Arxiv, 2023   Paper | Code

Optimization-inspired low-level vision architecture and results
Optimization-inspired learning with architecture augmentations and control mechanisms for low-level vision

Risheng Liu, Zhu Liu, Pan Mu, Xin Fan, Zhongxuan Luo

IEEE Transactions on Image Processing, 2023   Paper | Code

Perception-aware infrared-visible fusion under adversarial attacks
PAIF: Perception-Aware Infrared-Visible Image Fusion for Attack-Tolerant Semantic Segmentation

Zhu Liu, Jinyuan Liu, Benzhuang Zhang, Long Ma, Xin Fan, Risheng Liu

Proceedings of the 31st ACM International Conference on Multimedia, 2023   Paper | Code

Interactive image fusion and semantic segmentation framework
Multi-interactive Feature Learning and a Full-time Multi-modality Benchmark for Image Fusion and Segmentation

Jinyuan Liu, Zhu Liu, Guanyao Wu, Long Ma, Risheng Liu, Wei Zhong, Zhongxuan Luo, Xin Fan

Proceedings of the IEEE/CVF International Conference on Computer Vision, 2023   Paper | Code

Bi-level dynamic learning framework for multi-modality image fusion
Bi-level Dynamic Learning for Jointly Multi-modality Image Fusion and Beyond

Zhu Liu,Jinyuan Liu, Guanyao Wu, Long Ma, Xin Fan, Risheng Liu

Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, 2023   Paper | Code

Fast hierarchical fusion architecture qualitative result
Searching a hierarchically aggregated fusion architecture for fast multi-modality image fusion

Risheng Liu, Zhu Liu, Jinyuan Liu, Xin Fan

Proceedings of the 29th ACM International Conference on Multimedia, 2021   Paper | Code

Video deraining comparison with recovered scene details
Triple-level model inferred collaborative network architecture for video deraining

Pan Mu, Zhu Liu, Yaohua Liu, Risheng Liu, Xin Fan

IEEE Transactions on Image Processing, 2021   Paper | Code

Services

  • Program Committee: ICML, CVPR, ICCV, NeurIPS, IJCAI, AAAI, ACM MM, etc.
  • Reviewer: IEEE TIP, IEEE TCSVT, IEEE TMM, IEEE TIM, Information Fusion, etc.

Honors and Awards

  • Youth Student Fundamental Research Project for PHD students, NSFC,2025 (国家自然科学基金博士生项目)
  • Young Elite Scientists Sponsorship Program-Doctoral Student Special Plan, CAST,2025 (首批中国科协青年人才托举工程-博士生专项)
  • 2024 CSIG Outstanding Master's Thesis Incentive Program (CSIG 优硕激励, Top Ten Recipients)
  • Outstanding Ph.D. Dissertation Individual Scholarship of Dalian University of Technology, 2025
  • Outstanding Graduate, Liaoning Province, 2019, 2022, 2026
  • National Scholarship for Graduate Student in China, 2024
  • National Scholarship for Graduate Student in China, 2019 and 2021
  • National Scholarship in China, 2017 and 2018
  • Outstanding Master Dissertation Award of DLUT, 2022

TA Experience

  • Advanced Machine Learning, 2023-2026
  • Basics of C/C++ language programming, 2019, 2020