PhD – Neuronale 3D-Szenenrepräsentationen für den Fahrzeuginnenraum (w/m/div.)
Core
Researching neural 3D scene representations (NeRF, Gaussian Splatting, implicit Occupancy Models) for reliable reconstruction of passengers from 1-3 fixed camera perspectives in vehicle cabins.
Role type
PhD Researcher in 3D Computer Vision and Generative Modeling
Builds
End-to-End ADAS systems capable of inferring about passengers and surrounding traffic
Domain
Automotive AI, 3D Computer Vision, Generative Modeling
Deliverable
production ML models
Required skills
Machine Learning with PyTorch, Python programming, Computer Vision, 3D Geometry, Neural Scene Representation, Differentiable Rendering, Articulated Human Pose Estimation
Preferred skills
Experience with vision-based models, Knowledge of camera models and projective geometry, Familiarity with Multi-View Constraints
Technologies
PyTorch, NeRF, Gaussian Splatting, Implicit Occupancy Networks
Responsibilities
Adapt neural 3D scene representations to in-cabin perception challenges, Leverage known cabin geometry as integrated supervision, Integrate modern 3D-aware diffusion models as learned priors, Collaborate with industry experts to transfer research to safety-critical products
Seniority
PhD Candidate, Research & Development
