Research Scientist Intern, 3D Generative AI for Digital Humans (PhD)

The Reality Labs (RL) Research team brings together a world-class team of researchers, developers, and engineers to create the future of AR and VR. We are developing technologies to enable breakthrough AR glasses and VR headsets, working at the intersection of Computer Vision, Computer Graphics, and Machine Learning. The Virtual Humans team in Zurich, CH is expanding its focus to cutting-edge research in 3D Generative AI, with a particular emphasis on digital humans-encompassing neural representations and generative modeling of human shapes, meshes, and avatars.Research Topics: 3D generative modeling of digital humans/avatars. Both traditional mesh-based as well as gsplat representations. Applications of diffusion models, transformers, and large language models to 3D human modeling. Evaluation and benchmarking of generative 3D models for digital humans.* Integration of 3D generative models for digital humans into AR/VR pipelines.

Research Scientist Intern, 3D Generative AI for Digital Humans (PhD) Responsibilities:

• Prototyping, building, and analysis of novel 3D generative AI models for digital humans.
• Design and execution of experiments for 3D human shape generation, mesh synthesis, and related tasks.
• Collaboration with and support of other researchers across computer vision, graphics, and machine learning.
• Communication of research agenda, progress, and results.
• Potential to publish research in a top-tier conference/journal.

• Must obtain work authorization in the country of employment at the time of hire, and maintain ongoing work authorization during employment
• Strong understanding of machine learning, 3D geometry, optimization, and system building
• Experience with deep learning frameworks like PyTorch

• Intent to return to degree-program after the completion of the internship
• 1+ years experience in 3D generative modeling, mesh synthesis, or related areas for digital humans
• Familiarity with related papers such as 3DShape2VecSet, PolyGen, PolyDiff, etc
• Experience in cross-group and cross-culture collaboration
• Proven track record of achieving results as demonstrated in accepted papers at top vision, graphics, or AI conferences (CVPR, ECCV/ICCV, SIGGRAPH, NeurIPS, etc.)