Hi! I am a Lecturer (Assistant Professor) at Aston University. Previously, I was a Postdoctoral Research Associate at the University of Bristol, working with Prof. Nantheera Anantrasirichai (Pui) in the Visual Information Laboratory and the Bristol Vision Institute. Before that, I was a Senior Research Engineer at Baidu Inc. My research interests include image/video generation, 3D/4D reconstruction, and low-level vision.
If you are interested in collaborating, please contact me via email at g.huang2 [at] aston.ac.uk.
News
- 2026.10: BVI-RLV (low-light video dataset) is accepted to IEEE Transactions on Multimedia.
- 2026.09: OceanXL (large-scale underwater 3D Gaussian Splatting) is accepted to SIGGRAPH Asia 2026.
- 2026.09: Geometry beneath the Waves (sparse-view underwater 3DGS) is accepted as a SIGGRAPH Asia 2026 poster.
- 2026.05: Our review on underwater visual enhancement and 3D representation is published in Artificial Intelligence Review.
- 2026.04: Prune Wisely, Reconstruct Sharply (compact 3D Gaussian Splatting) is accepted to CVPR 2026.
- 2026.04: HDGS (sparse-view splatting via cascade depth loss) is accepted to ICME 2026.
- 2025.11: BEM (Bayesian neural networks for one-to-many image enhancement) is accepted to AAAI 2026.
- 2025.05: RUSplatting (sparse-view underwater scene reconstruction) is accepted to BMVC 2025.
- 2024.09: Layered Rendering Diffusion (controllable zero-shot image synthesis) is accepted to ECCV 2024.
- 2023.09: MIRL (masked image residual learning for deeper ViTs) is accepted to NeurIPS 2023.
- 2022.07: BQN (Busy-Quiet Net for action recognition) is published in IEEE TIP.
Publications
* Equal contribution
OceanXL: Large-scale Underwater 3D Gaussian Splatting via Block Partitioning and Adaptive Pruning
Haoran Wang, Shaoyu Cai, Adrian Azzarelli, Zhuodong Jiang, Guoxi Huang, Eng Tat Khoo, Brett Seymour, Fan Zhang, David Bull, Nantheera Anantrasirichai
SIGGRAPH Asia Conference Papers, 2026
A fast, scalable 3DGS framework that partitions large underwater scenes into spatially coherent blocks for memory-efficient reconstruction.
Geometry beneath the Waves: Dense Priors for Sparse-View Underwater 3D Gaussian Splatting
Harvey Caldeira, Haoran Wang, Guoxi Huang, Shaoyu Cai, Rachel Fu, Nantheera Anantrasirichai
SIGGRAPH Asia Posters, 2026
Adapts a feed-forward geometry foundation model to underwater degradation and uses its dense priors for sparse-view 3DGS.
BVI-RLV: A Fully Registered Dataset for Low-Light Video Enhancement
Ruirui Lin, Guoxi Huang†, Joanne Lin, Qi Sun, Alexandra Malyugina, David Bull, Nantheera Anantrasirichai†
IEEE Transactions on Multimedia (TMM), 2026
A fully registered low-light video dataset with benchmarks for low-light video enhancement.
Visual Enhancement and 3D Representation for Underwater Scenes: A Review
Guoxi Huang, Haoran Wang, Brett Seymour, Evan Kovacs, John Ellerbrock, Dave Blackham, Nantheera Anantrasirichai
Artificial Intelligence Review, 2026
A systematic review of underwater visual enhancement and underwater 3D reconstruction, from physical imaging models to learning-based methods.
Dynamic Weight-based Temporal Aggregation for Low-light Video Enhancement
Ruirui Lin, Guoxi Huang, Nantheera Anantrasirichai
IEEE International Conference on Image Processing (ICIP), 2026
DWTA-Net exploits long-term temporal cues through recurrent, dynamically weighted aggregation for low-light video enhancement under heavy noise.
Prune Wisely, Reconstruct Sharply: Compact 3D Gaussian Splatting via Adaptive Pruning and Difference-of-Gaussian Primitives
Haoran Wang, Guoxi Huang, Fan Zhang, David Bull, Nantheera Anantrasirichai
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026
A compact 3D Gaussian Splatting framework that combines adaptive pruning with Difference-of-Gaussian primitives.
Learning Fine-Grained Geometry for Sparse-View Splatting via Cascade Depth Loss
Wenjun Lu, Haodong Chen, Anqi Yi, Guoxi Huang, Yuk Ying Chung, Kun Hu, Zhiyong Wang
IEEE International Conference on Multimedia and Expo (ICME), 2026
HDGS uses cascaded multi-scale depth supervision to improve sparse-view splatting geometry and reconstruction fidelity.
Bayesian Neural Networks for One-to-Many Mapping in Image Enhancement
Guoxi Huang, Qirui Yang, Ruirui Lin, Zipeng Qi, David Bull, Nantheera Anantrasirichai
AAAI Conference on Artificial Intelligence (AAAI), 2026
An image enhancement model that can generate multiple enhanced images using Bayesian Neural Networks.
RUSplatting: Robust 3D Gaussian Splatting for Sparse-View Underwater Scene Reconstruction
Zhuodong Jiang, Haoran Wang, Guoxi Huang, Brett Seymour, Nantheera Anantrasirichai
British Machine Vision Conference (BMVC), 2025
A robust 3D Gaussian Splatting method for sparse-view underwater scene reconstruction.
Marine Snow Removal Using Internally Generated Pseudo Ground Truth
Alexandra Malyugina, Guoxi Huang, Eduardo Ruiz, Benjamin Leslie, Nantheera Anantrasirichai
European Signal Processing Conference (EUSIPCO), 2025
Generates paired snowy/clean training data from raw underwater videos to train a marine snow removal network.
BVI-Mamba: Video Enhancement Using a Visual State-Space Model for Low-Light and Underwater Environments
Guoxi Huang, Ruirui Lin, Yini Li, David Bull, Nantheera Anantrasirichai
Machine Learning from Challenging Data, 2025
A Visual State Space (Mamba) framework that reduces the memory and compute cost of low-light and underwater video enhancement.
Preprints
PixIE: Prompted Pixel-Space Low-Light Image Enhancement
Ruirui Lin, Guoxi Huang, David Bull, Nantheera Anantrasirichai
arXiv preprint, 2026
A feed-forward pixel-space low-light enhancement framework semantically prompted by foundation-model features.
Identity-Consistent Video Generation under Large Facial-Angle Variations
Bin Hu, Zipeng Qi, Guoxi Huang, Zunnan Xu, Ruicheng Zhang, Chongjie Ye, Jun Zhou, Xiu Li, Jingdong Wang
arXiv preprint, 2026
A multi-view conditioned reference-to-video framework that balances identity consistency and facial motion naturalness.
From Restoration to Reconstruction: Rethinking 3D Gaussian Splatting for Underwater Scenes
Guoxi Huang, Haoran Wang, Zipeng Qi, Wenjun Lu, David Bull, Nantheera Anantrasirichai
arXiv preprint, 2025
R-Splatting bridges underwater image restoration and 3DGS, fusing views from multiple restoration models into one reconstruction.
Semantic-guided Gaussian Splatting for High-Fidelity Underwater Scene Reconstruction
Zhuodong Jiang, Haoran Wang, Guoxi Huang, Brett Seymour, Nantheera Anantrasirichai
arXiv preprint, 2025
SWAGSplatting uses semantic guidance to balance 3DGS reconstruction across well-observed and degraded underwater regions.
Incrementally Learning Multiple Diverse Data Domains via Multi-Source Dynamic Expansion Model
Runqing Wu, Fei Ye, Qihe Liu, Guoxi Huang, Jinyu Guo, Rongyao Hu
arXiv preprint, 2025
MSDEM leverages multiple pre-trained backbones for continual learning over data from multiple distinct domains.