Ziming Liu

Research Scientist, Bosch Research, Shanghai · ziming.liu@cn.bosch.com

I work as a research scientist at Bosch Research in Shanghai, working on AI algorithms for autonomous driving and robotic applications. I obtained my PhD in the ACENTAURI team at Inria, France, advised by Dr. Philippe Martinet and Dr. Ezio Malis. Before that I received a master's degree in software engineering from Beijing Institute of Technology (北京理工大学), supervised by Prof. Guangyu Gao, co-supervised by Prof. Alex K. Qin and Dr. Lin Sun. My work focuses on AI planning for robotics, visual perception, visual localization, and video understanding.

News

Sep 2026HDVO published in IEEE Transactions on Intelligent Vehicles.
Aug 2026VectorReLoc accepted at ECCV 2026.

Publications

HDVO: Hybrid Dense Direct Visual Odometry

Ziming Liu (corresponding author), Ezio Malis, Philippe Martinet

IEEE Transactions on Intelligent Vehicles · 2026

Journal extension of the IROS 2022 and ITSC 2023 papers: stereo depth for metric scale, a robust Huber loss, and STC/LAM binary masks; real-time on Jetson Orin and Thor.

BibTeX

@article{HDVO-TIV-2026,
                                                                                                                                                        title     = {HDVO: Hybrid Dense Direct Visual Odometry},
                                                                                                                                                        author    = {Liu, Ziming and Malis, Ezio and Martinet, Philippe},
                                                                                                                                                        journal   = {IEEE Transactions on Intelligent Vehicles},
                                                                                                                                                        volume    = {11},
                                                                                                                                                        pages     = {930--945},
                                                                                                                                                        year      = {2026},
                                                                                                                                                        publisher = {IEEE},
                                                                                                                                                        doi       = {10.1109/TIV.2026.3708447}
                                                                                                                                                      }

Framework figure for VectorReLoc: vectorized SD map visual re-localization

VectorReLoc: Reliable Vectorized SD Map Visual Re-localization with Contrastive Feature Alignment

Ziming Liu (corresponding author), Quanjie Xiang, Yun Wang, Yiting Wang, Wen Chao, Zhuanjian Xu, Leichen Wang, Hao Sun, Guangyu Gao

ECCV 2026

Joint project between Bosch Research and the Bosch Cross-domain Computing Solutions (XC) division; deployed in Bosch automated driving products.

BibTeX

@inproceedings{vectorreloc-eccv-2026,
  title     = {VectorReLoc: Reliable Vectorized SD Map Visual Re-localization with Contrastive Feature Alignment},
  author    = {Liu, Ziming and Xiang, Quanjie and Wang, Yun and Wang, Yiting and Chao, Wen and Xu, Zhuanjian and Wang, Leichen and Sun, Hao and Gao, Guangyu},
  booktitle = {ECCV},
  year = {2026},
  publisher = {Springer},
  doi = {10.1007/978-3-032-37092-1_18}
}

Visual abstract of the Impromptu VLA dataset and benchmark results

Impromptu VLA: Open Weights and Open Data for Driving Vision-Language-Action Models

Haohan Chi, Huan-ang Gao, Ziming Liu (corresponding author), Jianing Liu, Chenyu Liu, Jinwei Li, Kaisen Yang, Yangcheng Yu, Zeda Wang, Wenyi Li, Leichen Wang, Xingtao Hu, Hao Sun, Hang Zhao, Hao Zhao

NeurIPS 2025

Outcome of an industry–academia collaboration between Bosch Research and Tsinghua University.

BibTeX

@inproceedings{chi2025impromptu,
  title     = {Impromptu VLA: Open Weights and Open Data for Driving Vision-Language-Action Models},
  author    = {Chi, Haohan and Gao, Huan-ang and Liu, Ziming and Liu, Jianing and Liu, Chenyu and Li, Jinwei and Yang, Kaisen and Yu, Yangcheng and Wang, Zeda and Li, Wenyi and Wang, Leichen and Hu, Xingtao and Sun, Hao and Zhao, Hang and Zhao, Hao},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year      = {2025}
}

Online map generation pipeline using the map query bank

Control Map Distribution using Map Query Bank for Online Map Generation

Ziming Liu, Leichen Wang, Ge Yang, Xinrun Li, Xingtao Hu, Hao Sun, Guangyu Gao

Preprint · April 2025

New state of the art on the OpenLane-V2 benchmark: 40.5% / 45.7% mAP on vehicle lane and pedestrian area.

arXiv
PDF

BibTeX

@misc{liu2025mqbank,
  title         = {Control Map Distribution using Map Query Bank for Online Map Generation},
  author        = {Liu, Ziming and Wang, Leichen and Yang, Ge and Li, Xinrun and Hu, Xingtao and Sun, Hao and Gao, Guangyu},
  year          = {2025},
  eprint        = {2504.03868},
  archivePrefix = {arXiv},
  primaryClass  = {cs.CV},
  doi           = {10.48550/arXiv.2504.03868}
}

One-stage deep stereo network

Ziming Liu, Ezio Malis, Philippe Martinet

ICASSP 2024 · April 2024

BibTeX

@inproceedings{stereoone-icassp-2024,
  title     = {One-stage deep stereo network},
  author    = {Liu, Ziming and Malis, Ezio and Martinet, Philippe},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  address   = {Seoul, South Korea},
  year      = {2024}
}

Multi-masks Generation for Increasing Robustness of Dense Direct Methods

Ziming Liu, Ezio Malis, Philippe Martinet

ITSC 2023 · September 2023

Extended in HDVO (IEEE T-IV 2026).

BibTeX

@inproceedings{MultiMasks-ITSC-2023,
  title        = {Multi-masks Generation for Increasing Robustness of Dense Direct Methods},
  author       = {Liu, Ziming and Malis, Ezio and Martinet, Philippe},
  booktitle    = {26th IEEE International Conference on Intelligent Transportation Systems (ITSC)},
  address      = {Bilbao, Spain},
  organization = {IEEE},
  month        = sep,
  year         = {2023}
}

Teaser figure for DANet: semi-supervised differentiated auxiliaries guided network

DANet: Semi-supervised differentiated auxiliaries guided network for video action recognition

Guangyu Gao, Ziming Liu (co-first author), Guangjun Zhang, Jinyang Li, A. K. Qin

Neural Networks · January 2023

An earlier self-supervised formulation of this work appeared as DTG-Net (preprint, 2020).

Publisher page

BibTeX

@article{DANet-NN-2023,
  title     = {DANet: Semi-supervised differentiated auxiliaries guided network for video action recognition},
  author    = {Gao, Guangyu and Liu, Ziming and Zhang, Guangjun and Li, Jinyang and Qin, A. K.},
  journal   = {Neural Networks},
  volume    = {158},
  pages     = {121--131},
  year      = {2023},
  publisher = {Elsevier}
}

A New Dense Hybrid Stereo Visual Odometry Approach

Ziming Liu, Ezio Malis, Philippe Martinet

IROS 2022 · October 2022

Extended in HDVO (IEEE T-IV 2026).

BibTeX

@inproceedings{HDVO-IROS-2022,
  title     = {A New Dense Hybrid Stereo Visual Odometry Approach},
  author    = {Liu, Ziming and Malis, Ezio and Martinet, Philippe},
  booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  address   = {Kyoto, Japan},
  publisher = {IEEE},
  pages     = {6998--7003},
  month     = oct,
  year      = {2022},
  doi       = {10.1109/IROS47612.2022.9981814}
}

HRDNet: High-resolution Detection Network for Small Objects

Ziming Liu, Guangyu Gao, Lin Sun, Zhiyuan Fang

ICME 2021 · oral

BibTeX

@inproceedings{liu2021hrdnet,
  title        = {HRDNet: high-resolution detection network for small objects},
  author       = {Liu, Ziming and Gao, Guangyu and Sun, Lin and Fang, Zhiyuan},
  booktitle    = {IEEE International Conference on Multimedia and Expo (ICME)},
  pages        = {1--6},
  year         = {2021},
  organization = {IEEE}
}

IPG-Net: Image Pyramid Guidance Network for Small Object Detection

Ziming Liu, Guangyu Gao, Lin Sun, Li Fang

CVPR Workshops · 2020

BibTeX

@inproceedings{liu2020ipg,
  title        = {IPG-Net: Image pyramid guidance network for small object detection},
  author       = {Liu, Ziming and Gao, Guangyu and Sun, Lin and Fang, Li},
  booktitle    = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
  pages        = {1026--1027},
  year         = {2020},
  organization = {IEEE}
}

Temporal Memory Network Towards Real-Time Video Understanding

Ziming Liu, Jinyang Li, Guangyu Gao, Alex K. Qin

IEEE Access · 2020

BibTeX

@article{liu2020temporal,
  title     = {Temporal Memory Network Towards Real-Time Video Understanding},
  author    = {Liu, Ziming and Li, Jinyang and Gao, Guangyu and Qin, Alex K.},
  journal   = {IEEE Access},
  volume    = {8},
  pages     = {223837--223847},
  year      = {2020},
  publisher = {IEEE}
}

DTG-Net: Differentiated Teachers Guided Self-Supervised Video Action Recognition

Ziming Liu, Guangyu Gao, A. K. Qin, Jinyang Li

Preprint · 2020

Extended and published as DANet in Neural Networks, 2023.

ResearchGate

BibTeX

@misc{DTG,
  title  = {DTG-Net: Differentiated Teachers Guided Self-Supervised Video Action Recognition},
  author = {Liu, Ziming and Gao, Guangyu and Qin, A. K. and Li, Jinyang},
  note   = {Preprint},
  year   = {2020}
}

Action Recognition with Bootstrapping based Long-range Temporal Context Attention

Ziming Liu, Guangyu Gao, A. K. Qin, Tong Wu, Chi Harold Liu

ACM MM 2019 · October 2019

PDF
ACM DL

BibTeX

@inproceedings{liu2019action,
  title        = {Action recognition with bootstrapping based long-range temporal context attention},
  author       = {Liu, Ziming and Gao, Guangyu and Qin, A. Kai and Wu, Tong and Liu, Chi Harold},
  booktitle    = {Proceedings of the 27th ACM International Conference on Multimedia (ACM MM)},
  pages        = {583--591},
  year         = {2019},
  organization = {ACM},
  doi          = {10.1145/3343031.3350916}
}

Workshop Challenges

ICCV 2019 VisDrone Challenge — 2nd place in the VID task.

Submitted as a team led by Ziming Liu (first author), with Jing Ge, Tong Wu and Guangyu Gao (supervisor). The reports below are the joint challenge summaries published by the organisers together with all participating teams.

Teaser figure for the VisDrone-DET2019 challenge report

VisDrone-DET2019: The Vision Meets Drone Object Detection in Image Challenge Results

Dawei Du, Pengfei Zhu, …, Ziming Liu

ICCV 2019 VisDrone Workshop

BibTeX

@inproceedings{dronedet,
  title        = {VisDrone-DET2019: The Vision Meets Drone Object Detection in Image Challenge Results},
  author       = {Du, Dawei and Zhu, Pengfei and Wen, Longyin and Bian, Xiao and Ling, Haibin and Hu, Qinghua and others},
  booktitle    = {IEEE/CVF International Conference on Computer Vision (ICCV) Workshops},
  month        = oct,
  year         = {2019},
  organization = {IEEE}
}

Teaser figure for the VisDrone-VID2019 challenge report

VisDrone-VID2019: The Vision Meets Drone Object Detection in Video Challenge Results

Pengfei Zhu, Dawei Du, …, Ziming Liu

ICCV 2019 VisDrone Workshop

BibTeX

@inproceedings{dronevid,
  title        = {VisDrone-VID2019: The Vision Meets Drone Object Detection in Video Challenge Results},
  author       = {Zhu, Pengfei and Du, Dawei and Wen, Longyin and Bian, Xiao and Ling, Haibin and Hu, Qinghua and others},
  booktitle    = {IEEE/CVF International Conference on Computer Vision (ICCV) Workshops},
  month        = oct,
  year         = {2019},
  organization = {IEEE}
}