Junwei Zheng

M.Sc. Junwei Zheng

  • Adenauerring 10
    Building 50.28, Room 106
    76131 Karlsruhe, Germany

About me

Welcome! I am a Ph.D. student at Computer Vision for Human-Computer Interaction Lab (CV:HCI), Karlsruhe Institute of Technology (KIT), Germany, under the supervision of Prof. Rainer Stiefelhagen and Prof. Kathrin Gerling. I received my B.Sc. degree from Guangdong University of Technology (GDUT) and M.Sc. degree from KIT.

My research topic focuses on assistive technology in the background of computer vision. More specifically, I dive into computer vision for scene understanding and embodied AI, especially vision-language navigation for people with visual impairments.

For KIT students who have passion for deep learning / computer vision / embodied AI / assistive technology and are looking for a bachelor / master thesis, please don’t hesitate to contact me with your transcript and curriculum vitae.

Topics include but are not limited to:

  • Scene understanding, e.g., semantic / instance / panoptic segmentation
  • Vision and language, e.g., vision-language navigation
  • Embodied AI
  • Assistive technology
  • Human computer interaction

For more information about me, please check my homepage.

Open thesis and Hiwi topics

  • ME2: Multi-Modal Enhanced Extendable Model for Vision-Language Navigation (Master Thesis, ongoing)


  • MateRobot: Material Recognition in Wearable Robotics for People with Visual Impairments  Junwei Zheng*, Jiaming Zhang*, Kailun Yang, Kunyu Peng, Rainer Stiefelhagen  preprint: arXiv  [paper] [code] [homepage
  • S3-MonoDETR: Supervised Shape&Scale-perceptive Deformable Transformer for Monocular 3D Object Detection  Xuan He, Kailun Yang, Junwei Zheng, Jin Yuan, Luis M. Bergasa, Hui Zhang, Zhiyong Li  preprint: arXiv  [paper] [code]
  • Unveiling the Hidden Realm: Self-supervised Skeleton-based Action Recognition in Occluded Environments  Yifei Chen, Kunyu Peng, Alina Roitberg, David Schneider, Jiaming Zhang, Junwei Zheng, Ruiping Liu, Yufan Chen, Kailun Yang, Rainer Stiefelhagen  preprint: arXiv  [paper] [code]
  • Elevating Skeleton-Based Action Recognition with Efficient Multi-Modality Self-Supervision  Yiping Wei, Kunyu Peng, Alina Roitberg, Jiaming Zhang, Junwei Zheng, Ruiping Liu, Yufan Chen, Kailun Yang, Rainer Stiefelhagen  preprint: arXiv  [paper] [code]
  • Attention-based Point Cloud Edge Sampling  Chengzhi Wu, Junwei Zheng, Julius Pfrommer, Juergen Beyerer  IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2023 (Highlight, Top 10%)  [paper] [code] [homepage]
  • Attention-based Part Assembly for 3D Volumetric Shape Modeling  Chengzhi Wu, Junwei Zheng, Julius Pfrommer, Juergen Beyerer  IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshop (CVPRW) 2023  [paper] [code]
  • Tightly-Coupled LiDAR-Visual SLAM Based on Geometric Features for Mobile Agents  Ke Cao, Ruiping Liu, Ze Wang, Kunyu Peng, Jiaming Zhang, Junwei Zheng, Zhifeng Teng, Kailun Yang, Rainer Stiefelhagen  IEEE/CVF International Conference on Robotics and Biomimetics (ROBIO) 2023  [paper]
  • Open Scene Understanding: Grounded Situation Recognition Meets Segment Anything for Helping People with Visual Impairments  Ruiping Liu, Jiaming Zhang, Kunyu Peng, Junwei Zheng, Ke Cao, Yufan Chen, Kailun Yang, Rainer Stiefelhagen  IEEE/CVF International Conference on Computer Vision Workshop (ICCVW) 2023  [paper] [code]