Ziyang Yan

I am a final-year phd student (January 2023 - ) in 3DOM: FBK and University of Trento , supervised by Prof. Fabio Remondino. I obtained my MSc degree from Hong Kong Baptist University and BSc degree from Guangdong University of Technology in China.

Email / Github / Google Scholar / LinkedIn

My research focuses on multimodal 3D reconstruction (e.g., Gaussian Splatting), 3D generation, and scene understanding. I am open to academic collaborations and communication at any time. Additionally, I am passionate about mentoring self-motivated junior students interested in advancing research in the 3D field. Feel free to reach out via email for inquiries or opportunities.

[2025.9.29] Our review paper is selected as the cover paper by Metrology!

[2025.9.18] We have one paper accepted by NIPS 2025!

[2025.7.5] Our work Eventvad has been accepted by ACM MM 2025!

[2025.6.25] Our work GM-MoE has been accepted by ICCV 2025!

[2025.5.12] Our work In-Context Meta LoRA has been accepted by IJCAI 2025!

[2025.4.03] I will visit CRCV at the University of Central Florida from June to December 2025, honored to be supervised by Prof. Mubarak Shah.

[2025.4.03] We have a review paper got accepted.

[2025.2.05] I will join BAAI as a research intern from February to June 2025 and am honored to work with Dr. Hao Zhao.

[2025.1.28] Our RenderWorld has been accepted by ICRA 2025!

[2023.10.01] Our NeRFBK has been accepted by GeoBench 2023 and the dataset is released!

Unifying Appearance Codes and Bilateral Grids for Driving Scene Gaussian Splatting
Nan Wang, Yuantao Chen, Lixing Xiao, Weiqing Xiao, Bohan Li, Zhaoxi Chen, Chongjie Ye, Shaocong Xu, Saining Zhang, Ziyang Yan, Pierre Merriaux, Lei Lei, Tianfan Xue, Hao Zhao*
(*indicates corresponding author)
Annual Conference on Neural Information Processing Systems (NeurIPS 2025)
[Paper] [Project] [Code]

Eventvad: Training-free event-aware video anomaly detection
Yihua Shao, Haojin He, Sijie Li, Siyu Chen, Xinwei Long, Fanhu Zeng, Yuxuan Fan, Muyang Zhang, Ziyang Yan, Ao Ma, Xiaochen Wang, Hao Tang, Yan Wang, Shuyan Li*
(*indicates corresponding author)
ACM Multimedia (MM 2025)
[Paper] [Code] [机器之心]

GM-MoE: Low-Light Enhancement with Gated-Mechanism Mixture-of-Experts
Minwen Liao, Hao Bo Dong, Xinyi Wang, Ziyang Yan*, Yihua Shao*
(*indicates corresponding author)
International Conference of Computer Vision (ICCV 2025)
[Paper]

Learning-Based 3D Reconstruction Methods for Non-Collaborative Surfaces—A Metrological Evaluation
Ziyang Yan*, Nazanin Padkan, Paweł Trybała, Elisa Mariarosaria Farella, Fabio Remondino
(*indicates corresponding author)
Metrology (journal cover) [Paper]

AccidentBlip: Agent of Accident Warning based on MA-former
Yihua Shao, Yeling Xu, Xinwei Long, Siyu Chen, Ziyang Yan, Yang Yang, Haoting Liu, Yan Wang, Hao Tang, Zhen Lei*
(*indicates corresponding author)
IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV 2025)
[Paper]

In-Context Meta LoRA Generation
Yihua Shao, Minxi Yan, Yang Liu, Siyu Chen, Wenjie Chen, Xinwei Long, Ziyang Yan, Lei Li, Chenyu Zhang, Nicu Sebe, Hao Tang, Yan Wang, Hao Zhao, Mengzhu Wang, Jingcai Guo*
(*indicates corresponding author)
The 34th International Joint Conference on Artificial Intelligence (IJCAI 2025)
[Paper]

3DSceneEditor: Controllable 3D Scene Editing with Gaussian Splatting
Ziyang Yan, Lei Li*, Yihua Shao, Siyu Chen, Wuzong Kai, Jenq-Neng Hwang, Hao Zhao, Fabio Remondino*
(*indicates corresponding author)
arXiv
[Paper] [Code] [Project]

GWQ: Gradient-Aware Weight Quantization for Large Language Models
Yihua Shao, Siyu Liang, Zijian Ling, Minxi Yan, Haiyang Liu, Siyu Chen, Ziyang Yan, Chenyu Zhang, Haotong Qin, Michele Magno, Yang Yang, Zhen Lei, Yan Wang, Jingcai Guo, Ling Shao, Hao Tang
arXiv
[Paper]

RenderWorld: World Model with Self-Supervised 3D Label
Ziyang Yan*, Wenzhen Dong*, Yihua Shao*, Yuhang Lu, Liu Haiyang, Jingwen Liu, Haozhe Wang, Zhe Wang, Yan Wang, Fabio Remondino, Yuexin Ma**
(*indicates equal contribution, **indicates corresponding author)
2025 IEEE International Conference on Robotics and Automation (ICRA 2025)
[Paper]

NeRFBK: a holistic dataset for benchmarking NeRF-based 3D reconstruction
Ziyang Yan, Gabriele Mazzacca, Simone Rigon, Elisa Mariarosaria Farella, Pawel Trybala, Fabio Remondino*
(*indicates corresponding author)
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
[Paper] [Dataset]

A critical analysis of nerf-based 3d reconstruction
Fabio Remondino, Ali Karami, Ziyang Yan, Gabriele Mazzacca, Simone Rigon, Rongjun Qin

Remote Sensing (IF=4.2)
[Paper]

2017 Mathematical Contest In Modeling
Huang Rui, Yan Ziyang, Zhan Zhiyong

Third Prize

We built a Urban Smart Growth Evaluation System and estimated the development of the two cities, Jinchang, Gansu and Sacramento, California by three different ranking indicators with the AHP method. Next, we proposed the smart growth policy of the two cities based on the evaluation result. After that, we set up a Comprehensive Evaluation Model attached with both effectiveness and feasibility to rank the initiatives mentioned in our smart growth policy. For the feasibility, we adopted the Fuzzy Comprehensive Evaluation System to quantify it. Finally, we won third prize in this contest.

2017 Challenge Cup Competition of Science Achievement in China
J.D Chen, Z.X Zhang, Z.B Yao, Z.H Shi, Z.Y Yan

Second Prize

We designed an electric inspection UAV based on the M100 aircraft in this competition. We used STM32 and A53 embedded system to position the dangerous objects on the transmission line through image processing. We won second prize in the end. (Top 20%)

2016 International Sensor Innovation and Entrepreneurship Competition (South China Division)
Z.Y Yan, J,D Chen, Z.B Yao J,X Zhang

Second Prize

We designed a new 'flying wing' aircraft based on flying wing X5. The product is integrated traditional designed style of flying wing and Cessna to further increases the aspect ratio and vertical and horizontal stability. Our product successfully increased the wind resistance, stability, mounting load and maximum range and also reduce the wing load. We got second prize in this competition.

2015 Education Robot contest of China (ERCC)
X Chen, Z.Y Yan, H,R Zhang

Second Prize

The competition requires us to design a robot that can move according to the black lines on the ground.There are multiple stations on the ground. When the robot passes through the stations, it need to read the station name through the RFID module and correctly identify the station name. We got second prize in this competition.

Last update: 2025.05.12. Thanks Jon Barron and QingYong HU