I am currently a doctoral student at the Graduate School of Information Science and Technology (IST), The University of Tokyo, with Prof. Yinqiang Zheng.
Prior to this, I obtained my M.Sc. degree at the School of Computer Science, Peking University, under the supervision of Prof. Yisong Chen. I worked at Graphics and Interaction Lab (GIL) led by Prof. Guoping Wang on a scalable 3D reconstruction system (i23D) especially for modeling urban scenes from massive UAV imagery.
My research interests mainly lie in the scope of computer vision and computer graphics. More specifically, I study the algorithms of 3D reconstruction (photogrammetry), neural rendering, implicit representations, and computational photography. I also have experience in remote sensing, visual localization (VPS), geometry processing, and graph signal processing.
I am able to speak Mandarin (native), English (IELTS 8.0), Spanish (DELE B1), Portuguese (CAPLE CIPLE), and now struggling with Japanese (JLPT N3) XD.
Should you have any questions, please feel free to reach out!
🔥 News
- 2024.09: 🥇 Our team ranks 1st in ECCV 2024 GigaRendering Challenge.
- 2024.08: 🎉 One paper is accepted by ECCV 2024 as an oral presentation. See you at Milan.
- 2024.05: 🥇 Our team ranks 1st in ScanNet++ Novel View Synthesis Challenge of CVPR 2024.
- 2023.10: 🇯🇵 I start pursuing my Ph.D. degree at UTokyo as a Todai Fellowship student.
See Previous
- 2023.06: 🥇 Our team ranks 1st in GAIIC 2023 GigaRendering Challenge.
- 2023.05: I have passed the defense of my master thesis, titled Semantic Reconstruction of 3D Models for Urban Scenes.
- 2023.02: 🥇 Our team ranks 1st in GigaReconstruction Challenge of GigaVision 2022.
- 2022.10: I receive Benz Scholarship offered by Daimler Greater China Ltd.
- 2022.09: 🎉 One paper is accepted by BMVC 2022.
- 2022.06: 🎉🎉 Two papers are accepted by ECCV 2022.
- 2022.04: 🎉 One paper is accepted by TVCG.
📃 Publications
* = equal contributions
🌟 Selected Publications
RPBG: Towards Robust Neural Point-based Graphics in the Wild
Qingtian Zhu, Zizhuang Wei, Zhongtian Zheng, Yifan Zhan, Zhuyu Yao, Jiawang Zhang, Kejian Wu, Yinqiang Zheng
[arxiv] [code]
- We propose a robust alternative for NVS that performs neural re-rendering of triangulated points from posed images.
- It stably outperforms radiance-field-based methods especially in convergence time, perceptual quality, and generalizability across diverse scene types.
Sobolev Training for Implicit Neural Representations with Approximated Image Derivatives
Wentao Yuan, Qingtian Zhu, Xiangyue Liu, Yikang Ding, Haotian Zhang, Chi Zhang
[arxiv] [paper] [code]
- We propose and evaluate a new training paradigm for INRs where numerical derivatives are used to supervise analytical derivatives of INRs in addition to signal values.
- Approximation performance is improved with a notable margin on tasks such as image regression, inverse rendering and audio regression.
TransMVSNet: Global Context-aware Multi-view Stereo Network with Transformers
Yikang Ding*, Wentao Yuan*, Qingtian Zhu, Haotian Zhang, Xiangyue Liu, Yuanjiang Wang, Xiao Liu
[arxiv] [paper] [code]
- Transformers are leveraged to extract deep features within and across images and capture global contexts.
- TransMVSNet ranks 1st on all MVS datasets, including DTU, BlendedMVS and Tanks and Temples.
📁 Full List
2024
-
RPBG: Towards Robust Neural Point-based Graphics in the Wild
Qingtian Zhu, Zizhuang Wei, Zhongtian Zheng, Yifan Zhan, Zhuyu Yao, Jiawang Zhang, Kejian Wu, Yinqiang Zheng
ECCV 2024 (Oral) [arxiv] [code] -
Learning-based Multi-view Stereo: A Survey
Fangjinhua Wang*, Qingtian Zhu*, Di Chang*, Quankai Gao, Junlin Han, Tong Zhang, Richard Hartley, Marc Pollefeys
[arxiv] -
ToMiE: Towards Modular Growth in Enhanced SMPL Skeleton for 3D Human Gaussians with Animatable Garments
Yifan Zhan, Qingtian Zhu, Muyao Niu, Mingze Ma, Jiancheng Zhao, Zhihang Zhong, Xiao Sun, Yu Qiao, Yinqiang Zheng
[arxiv]
2022
-
Bidirectional Hybrid LSTM Based Recurrent Neural Network for Multi-view Stereo
Zizhuang Wei, Qingtian Zhu, Chen Min, Yisong Chen, Guoping Wang
TVCG [paper] -
TransMVSNet: Global Context-aware Multi-view Stereo Network with Transformers
Yikang Ding*, Wentao Yuan*, Qingtian Zhu, Haotian Zhang, Xiangyue Liu, Yuanjiang Wang, Xiao Liu
CVPR 2022 [arxiv] [paper] [code] -
Sobolev Training for Implicit Neural Representations with Approximated Image Derivatives
Wentao Yuan, Qingtian Zhu, Xiangyue Liu, Yikang Ding, Haotian Zhang, Chi Zhang
ECCV 2022 [arxiv] [paper] [code] -
KD-MVS: Knowledge Distillation Based Self-supervised Learning for Multi-view Stereo
Yikang Ding, Qingtian Zhu, Xiangyue Liu, Wentao Yuan, Haotian Zhang, Chi Zhang
ECCV 2022 [arxiv] [paper] [code] -
Hybrid Cost Volume Regularization for Memory-efficient Multi-view Stereo Networks
Qingtian Zhu, Zizhuang Wei, Zhongtao Wang, Yisong Chen, Guoping Wang
BMVC 2022 [paper]
2021
-
MegLoc: A Robust and Accurate Visual Localization Pipeline
Shuxue Peng*, Zihang He*, Haotian Zhang*, Ran Yan*, Chuting Wang*, Qingtian Zhu*, Xiao Liu
[arxiv] -
AA-RMVSNet: Adaptive Aggregation Recurrent Multi-view Stereo Network
Zizhuang Wei, Qingtian Zhu, Chen Min, Yisong Chen, Guoping Wang
ICCV 2021 [arxiv] [paper] [code] -
Deep Learning for Multi-view Stereo via Plane Sweep: A Survey
Qingtian Zhu, Chen Min, Zizhuang Wei, Yisong Chen, Guoping Wang
[arxiv]
2019
- Efficient Multi-class Semantic Segmentation of High Resolution Aerial Imagery with Dilated LinkNet
Qingtian Zhu, Yumin Zheng, Yulai Jiang, Junli Yang
IGARSS 2019 (Oral) [paper] [code]
🏆 Awards
- 2024.09: Winner in ECCV 2024 GigaRendering Challenge
- 2024.05: Winner in Novel View Synthesis Challenge of ScanNet++ CVPR 2024 Workshop [website]
- 2023.09: The University of Tokyo Fellowship (Todai Fellowship)
- 2023.06: Winner in GAIIC 2023 GigaRendering Challenge [report]
- 2023.02: Winner in GigaReconstruction Challenge of GigaVision 2022 [report]
- 2023.02: Runner-up in GigaRendering Challenge of GigaVision 2022 [report]
- 2022.10: Benz Scholarship
- 2021.10: Winner in Indoor and Outdoor Visual Localization Challenge of ICCV 2021 Workshop on Long-term Visual Localization under Changing Conditions [website] [talk] [report]
- 2021.09: PKU Special Academic Scholarship
- 2020.06: Outstanding Graduate and Outstanding Thesis
- 2019.06: Winner in Visual Localization Challenge of CVPR 2019 Workshop on Long-term Visual Localization under Changing Conditions [website]
- 2017.11: National Scholarship
🎓 Educations
- 2023.10 - present:
Ph.D. in Mechano-Informatics, The University of Tokyo - 2020.09 - 2023.07:
M.Sc. in Computer Software and Theory, Peking University - 2016.09 - 2020.06:
B.Eng. in Internet of Things Engineering, Beijing University of Posts and Telecommunications - 2016.09 - 2020.06:
B.S.E. (First Class Honour) in Internet of Things Engineering, Queen Mary University of London
💻 Internships
- 2023.05 - 2024.03:
XREAL (Nreal), supervised by Zhuyu Yao, Jiawang Zhang, and Kejian Wu - 2022.02 - 2022.07:
XR Lab, Alibaba DAMO Academy, supervised by Lingzhi Li and Li Shen - 2021.08 - 2022.02:
MEGVII Research, supervised by Ran Yan, Haotian Zhang, and Xiao Liu - 2019.03 - 2019.06:
NLPR, CASIA, supervised by Shuhan Shen