Yu Sun Data-verified
Affiliation confirmed via AI analysis of OpenAlex, ORCID, and web sources.
Assistant Professor
faculty
Research Areas
Links
Biography and Research Information
OverviewAI-generated summary
Yu Sun's research focuses on the application of machine learning and deep learning techniques to solve complex problems in areas such as medical imaging, robotics, and financial forecasting. Sun has explored coordinate-based internal learning for tomographic imaging and developed methods for recovering 3D refractive index maps from limited measurements using neural fields. Additionally, Sun has investigated scalable plug-and-play ADMM algorithms with convergence guarantees and efficient model-based deep learning with theoretical guarantees.
Further research interests include path planning algorithms for robot arms, utilizing improved RRT* and BP neural networks. Sun has also applied machine learning to financial markets, developing LSTM-XGBoost models for stock price forecasting optimized with Bayesian methods. Other work involves video coding optimization and enhancing aircraft detection in remote sensing imagery using progressive class-aware instance enhancement.
With an h-index of 21 and over 116 publications, Sun is recognized as a highly cited researcher. Sun maintains an active lab website, indicating ongoing research endeavors.
Metrics
- h-index: 22
- Publications: 109
- Citations: 3,698
Selected Publications
-
LDG-PCGC: Lossless Dynamically Grouped Point Cloud Geometry Compression (2026)
-
LDG-PCGC: Lossless Dynamically Grouped Point Cloud Geometry Compression (2026)
-
A Multi-Layer End-to-End 360 Image Compression (2025)
-
Channel and space-based joint rate allocation algorithm (2025)
-
Fast CU Partition Algorithm For 360-Degree Videos on VVC (2025)
-
A Novel Mode Selection-Based Fast Intra Prediction Algorithm for Spatial SHVC (2023)
-
A Probability-Based Zero-Block Early Termination Algorithm for QSHVC (2023)
-
Hybrid Strategies for Efficient Intra Prediction in Spatial SHVC (2022)
Collaboration Network
Top Collaborators
- CoIL: Coordinate-Based Internal Learning for Tomographic Imaging
- Recovery of continuous 3D refractive index maps from discrete intensity-only measurements using neural fields
- Scalable Plug-and-Play ADMM With Convergence Guarantees
- SGD-Net: Efficient Model-Based Deep Learning With Theoretical Guarantees
- CoIL: Coordinate-based Internal Learning for Imaging Inverse Problems
Showing 5 of 14 shared publications
- CoIL: Coordinate-Based Internal Learning for Tomographic Imaging
- SGD-Net: Efficient Model-Based Deep Learning With Theoretical Guarantees
- CoIL: Coordinate-based Internal Learning for Imaging Inverse Problems
- Deformation-Compensated Learning for Image Reconstruction Without Ground Truth
- Deep Image Reconstruction Using Unregistered Measurements Without Groundtruth
Showing 5 of 9 shared publications
- CoIL: Coordinate-Based Internal Learning for Tomographic Imaging
- Scalable Plug-and-Play ADMM With Convergence Guarantees
- SGD-Net: Efficient Model-Based Deep Learning With Theoretical Guarantees
- CoIL: Coordinate-based Internal Learning for Imaging Inverse Problems
- Coordinate-Based Seismic Interpolation in Irregular Land Survey: A Deep Internal Learning Approach
Showing 5 of 8 shared publications
- SGD-Net: Efficient Model-Based Deep Learning With Theoretical Guarantees
- Deformation-Compensated Learning for Image Reconstruction Without Ground Truth
- Deep Image Reconstruction Using Unregistered Measurements Without Groundtruth
- Joint Reconstruction and Calibration Using Regularization by Denoising with Application to Computed Tomography
- Stochastic Deep Unfolding for Imaging Inverse Problems
Showing 5 of 6 shared publications
- Scalable Plug-and-Play ADMM With Convergence Guarantees
- SGD-Net: Efficient Model-Based Deep Learning With Theoretical Guarantees
- Stochastic Deep Unfolding for Imaging Inverse Problems
- CoIL: Coordinate-Based Internal Learning for Tomographic Imaging
- CoIL: Coordinate-based Internal Learning for Imaging Inverse Problems
- Joint Reconstruction and Calibration Using Regularization by Denoising with Application to Computed Tomography
- Scalable Plug-and-Play ADMM With Convergence Guarantees
- Provable Probabilistic Imaging Using Score-Based Generative Priors
- Scalable image reconstruction in optical tomography using deep priors
- Recovery of continuous 3D refractive index maps from discrete intensity-only measurements using neural fields
- Zero-Shot Learning of Continuous 3D Refractive Index Maps from Discrete Intensity-Only Measurements
- Scalable image reconstruction in optical tomography using deep priors
- Deformation-Compensated Learning for Image Reconstruction Without Ground Truth
- Deep Image Reconstruction Using Unregistered Measurements Without Groundtruth
- MoDIR: Motion-Compensated Training for Deep Image Reconstruction without Ground Truth
- Deformation-Compensated Learning for Image Reconstruction Without Ground Truth
- Deep Image Reconstruction Using Unregistered Measurements Without Groundtruth
- MoDIR: Motion-Compensated Training for Deep Image Reconstruction without Ground Truth
- Probability-Based Fast Intra Prediction Algorithm for Spatial SHVC
- Gaussian Distribution-based Mode Selection for Intra Prediction of Spatial SHVC
- Fast CU Partition Algorithm For 360-Degree Videos on VVC
- Probability-Based Fast Intra Prediction Algorithm for Spatial SHVC
- Gaussian Distribution-based Mode Selection for Intra Prediction of Spatial SHVC
- Fast CU Partition Algorithm For 360-Degree Videos on VVC
- Probability-Based Fast Intra Prediction Algorithm for Spatial SHVC
- Gaussian Distribution-based Mode Selection for Intra Prediction of Spatial SHVC
- Recovery of continuous 3D refractive index maps from discrete intensity-only measurements using neural fields
- Zero-Shot Learning of Continuous 3D Refractive Index Maps from Discrete Intensity-Only Measurements
- Recovery of continuous 3D refractive index maps from discrete intensity-only measurements using neural fields
- Zero-Shot Learning of Continuous 3D Refractive Index Maps from Discrete Intensity-Only Measurements
Similar Researchers
Based on overlapping research topics