Wentian Gan 甘文天

School of Geodesy and Geomatics, Wuhan University

Mail: gwt2019@whu.edu.cn

About Me

My research focuses on computer vision, specifically on 3D reconstruction and photogrammetry , I aim to integrate traditional 3D reconstruction methods with innovative deep learning techniques to address challenges in robustness and computational efficiency. Currently, I am developing an efficient and robust real-time Structure from Motion (SfM) system for high-precision 3D reconstruction and reliable camera pose estimation, with applications in autonomous driving, unmanned navigation, and emergency disaster response. Also, I am exploring cutting-edge deep learning methods for 3D reconstruction, such as 3D Gaussian Splatting and VGGT , to enhance the quality and speed of 3D modeling from images and videos.

News

[2024.3]: 🎉 One paper is accepted by ISPRS TCI Mid-term Symposium.
[2024.8]: 🎉 One paper is accepted by ECCV 2024.
[2025.9]: 🎉 One WhitePaper is accepted by IEEE GRSM.

Publications

Joint first authors are indicated using * and corresponding authors are indicated using †.

2024

Efficient Feature Matching and Pose-graph Initialization for SfM
Wen Tian Gan, Zong Qian Zhan, and Xin Wang†
ISPRS TCI Mid-term Symposium, 2024, Poster

LVG-SfM: Learning-Based View-Graph Generation for Robust on-the-Fly SfM
Wentian Gan*, Yifei Yu*, Giulio Perda, Luca Morelli, Rui Xia, Zongqian Zhan, Xin Wang† & Fabio Remondino
European Conference on Computer Vision (ECCV), 2024, Oral

2025

SfM on-the-fly: A robust near real-time SfM for spatiotemporally disordered high-resolution imagery from multiple agents
Zongqian Zhan, Yifei Yu, Rui Xia , Wentian Gan , Hong Xie , Giulio Perda , Luca Morelli , Fabio Remondino , Xin Wang†
ISPRS Journal of Photogrammetry and Remote Sensing, 2025
Website / Code

Gaussian On-the-Fly Splatting: A Progressive Framework for Robust Near Real-Time 3DGS Optimization
Y Xu, Y Yu, W Gan, T Wang, Z Zhan, H Cheng, X Wang†
IEEE Robotics and Automation Letters(RA-L), Accepted/InPress, 2025

A-TDOM: Active TDOM via On-the-Fly 3DGS
Yiwei Xu, Xiang Wang, Yifei Yu, Wentian Gan, Luca Morelli, Giulio Perda, Xiongwu Xiao, Zongqian Zhan, Xin Wang†, and Fabio Remondino
IEEE International Journal of Computer Vision(IJCV), UnderReview, 2025