Dong In Lee (이동인)
Visiting scholar @ Purdue Univ | Ph.D. student @ Korea Univ
dilee99@korea.ac.kr
Welcome to my website! I am a Ph.D. student in the Computer Vision Lab (KUAICV) at Korea University, advised by Prof. Sangpil Kim.
I am currently a visiting scholar at Purdue University, collaborating with Prof. Karthik Ramani at the Convergence Design Lab.
My research interests lie in 3D computer vision, especially in:
- 3D/4D scene editing and reconstruction
- Human-object interaction
- AI watermarking
Outside of research, I enjoy playing soccer, working out, and traveling. I also love learning new things beyond my research area. I’m also a big fan of Arsenal.
If you’re interested in my research and collaboration, feel free to connect!
news
| Jul 04, 2025 | Our paper on 3D human-object interaction has been accepted to ACM MM 2025. |
|---|---|
| Mar 10, 2025 | Two of our papers — EditSplat on 3D scene editing, and IKNE on 3D scene segmentation — have been accepted to CVPR 2025. |
| Feb 20, 2025 | I have started my journey as a visiting scholar at Purdue University🇺🇸! |
| Mar 06, 2024 | Our paper on 3D Watermarking, WateRF, was accepted to CVPR 2024. |
| Sep 01, 2023 | Started the integrated MS/PhD program in the Computer Vision Lab (KUAICV), Korea University. |
selected publications
- ACM MM 2025
Temporally Consistent Amodal Completion for 3D Human-Object Interaction ReconstructionProceedings of the 33nd ACM International Conference on Multimedia. 2025, 2025 - CVPR 2025
Editsplat: Multi-view fusion and attention-guided optimization for view-consistent 3d scene editing with 3d gaussian splattingIn Proceedings of the Computer Vision and Pattern Recognition Conference, 2025 - CVPR 2025
Insightful Instance Features for 3D Instance SegmentationIn Proceedings of the Computer Vision and Pattern Recognition Conference, 2025 - CVPR 2024
Waterf: Robust watermarks in radiance fields for protection of copyrightsIn Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024 - Applied Sciences
Crop disease diagnosis with deep learning-based image captioning and object detectionApplied Sciences, 2023 - ICCE 2024
Development of Parameter Bit Operation Framework for Deep Neural NetworksIn 2024 IEEE International Conference on Consumer Electronics (ICCE), 2024