Home

Jihong Ju

I bring innovative research to cutting-edge AR/VR products. In particular, my work focuses on real-time 3D reconstruction and spatial understanding using edge computing. Hope to see the virtual and physical 3D worlds seamlessly blended for an even better immersive experience!

Currently, I am a Computer Vision Engineer at Qualcomm XR Labs, where I work on advancing 3D reconstruction and understanding technologies for AR/VR glasses. I earned my M.Sc. in Computer Science from TU Delft in 2017. For more information about my work and background, feel free to explore my profile.

My research interests lie in 3D Computer Vision and Simultaneous Localization and Mapping (SLAM). Specifically, I am exploring multi-modal perception systems for applications in AR/VR, autonomous driving, and robotics. Below is a list of selected publications.

• Ju, J., Tseng, C. W., Bailo, O., Dikov, G., & Ghafoorian, M. (2023). Dg-recon: Depth-guided neural 3d scene reconstruction. In ICCV.
• Langer, F., Ju, J., Dikov, G., Reitmayr, G., & Ghafoorian, M. (2024). FastCAD: Real-Time CAD Retrieval and Alignment from Scans and Videos. In ECCV.
• Nguyen, D. K., Ju, J., Booij, O., Oswald, M. R., & Snoek, C. G. (2022). Boxer: Box-attention for 2d and 3d transformers. In CVPR.
• Corbato, C. H., Bharatheesha, M., Van Egmond, J., Ju, J., & Wisse, M. (2018). Integrating different levels of automation: Lessons from winning the amazon robotics challenge 2016. IEEE Trans. Ind. Inform.

In my spare time, I enjoy expanding my knowledge. Some of the insights are shared as blog posts and medium articles. I was also actively involved in open-source contributions. Feel free to explore my Github repositories.