CV
Experience
Research Assistant — DFKI, Augmented Vision Department, Germany February 2024 – Present
- Developed ReLaGS, a framework for open-vocabulary 3D scene understanding using hierarchical language-distilled Gaussian scenes and 3D semantic scene graphs; accepted at CVPR 2026.
- Built an open-vocabulary 3D scene understanding system with Gaussian Splatting, achieving +4% IoU improvement on segmentation benchmarks.
- Demonstrated semantic 3D object retrieval in complex real-world scenes.
- Improved NeRF rendering speed by integrating MSI, enabling real-time VR applications.
Software Engineer — TRU, Egypt December 2022 – January 2024
- Developed and optimized the backend for a large-scale e-commerce platform using Django and GraphQL.
- Improved notification reliability and scalability while maintaining clean architecture (SOLID principles).
Research Intern — Microsoft ATL, Egypt July 2022 – September 2022
- Investigated parameter averaging techniques for model enhancement in Neural Machine Translation and Speech Recognition.
Research Intern — DFKI, Germany April 2022 – June 2022
- Reduced data loading time for a 3TB dataset from 3 days to 1 hour using multithreading and TAR compression.
- Implemented a GAN-based sign language production system; co-authored SynthSL (IEEE FG 2024).
Education
M.Sc. in Computer Science — RPTU Kaiserslautern-Landau October 2023 – January 2026 (expected)
- GPA: 1.0/1.0 (with distinction)
- Thesis: Open-Vocabulary 3D Scene Understanding with Gaussian Splatting
B.Sc. in Computer Science and Engineering — German University in Cairo September 2018 – July 2023
- GPA: 0.77/0.7 — Ranked 7th out of 350+ students
- Thesis: Sign Language Production using GANs
Publications
ReLaGS: Relational Language Gaussian Splatting Yaxu Xie*, Abdalla Arafa*, Alireza Javanmardi, Christen Millerdurai, Jia Cheng Hu, Shaoxiang Wang, Alain Pagani, Didier Stricker CVPR 2026 · *Equal contribution
SynthSL: Expressive Humans for Sign Language Image Synthesis Jilliam M. Díaz Barros*, Chen-Yu Wang*, Jameel Malik, Abdalla Arafa, Didier Stricker IEEE FG 2024 · *Equal contribution
Scholarships & Awards
| Award | Year |
|---|---|
| Full Merit-Based Scholarship — RPTU Kaiserslautern-Landau | 2023–2026 |
| DAAD Scholarship — Academic Excellence | 2019 |
| Full Merit-Based Scholarship — German University in Cairo | 2018–2023 |
| National Academic Ranking: 7th in Egypt | 2018 |
Technical Skills
Computer Vision & Graphics: Gaussian Splatting, NeRF, Neural Rendering, OpenGL
Machine Learning & AI: Deep Learning, GANs, Transformers, HuggingFace, Scikit-Learn
Frameworks & Libraries: PyTorch, OpenCV, Django, Spring Boot
Programming: Python, C++, Java, JavaScript
Tools & Platforms: Git, Docker, Linux, SLURM
Data Science: NumPy, Pandas, Matplotlib