Internship 3D Scene Editing and Generation

About the role

During a six‑month internship you will contribute to a cutting‑edge 3D scene editing pipeline that leverages 3D Gaussian Splatting, video diffusion models and semantic embeddings such as CLIP and DINO. The work supports robotics and autonomous‑driving research and is implemented in PyTorch. Successful interns are expected to continue the project as a master thesis.

Key responsibilities

• Develop and refine the 3D scene editing pipeline using 3D Gaussian Splatting and video diffusion techniques.
• Integrate semantic embedding models (CLIP, DINO) to enhance perception for robotics and autonomous‑driving applications.

Required profile

• Enrolled in a Master program in Computer Science, Electrical Engineering, Mechatronics or a related field.
• Strong ability to work independently and in a structured manner, with good English communication skills.

Required skills

• Python programming.
• Experience with deep‑learning frameworks (PyTorch, TensorFlow).
• Knowledge of diffusion models and 3D Gaussian Splatting.
• Familiarity with semantic embedding models such as CLIP and DINO.