The Reality Labs (RL) Research Team in Zurich, CH is looking for research interns to tackle the most ambitious problems in the creation of photorealistic digital humans. In this internship, we aim to build photorealistic human avatars from limited observations by leveraging generative models as a strong prior for realistic human appearance. We plan to augment pre-trained generative models with additional controls over head pose, face expressions and hair motion, and work towards generation of 4D-consistent photorealistic avatars. We envision a world where people connect across vast distances so our roles focus on capturing and modeling highly detailed and realistic human appearance. As part of our team you will explore new ways of generating controllable photorealistic avatars and will be able to leverage high-quality 4D captures, to go beyond existing state-of-the-art approaches.Our internships are sixteen (16) to twenty four (24) weeks long and we have various start dates throughout the year.RESEARCH TOPICSVirtual humans, animatable avatar creation.Controllable image and video generation.*Neural representations for shape, motion and appearance.

• Develop novel algorithms in areas such as controllable avatar generation, realistic avatar reconstruction from limited observations, and temporally- and view-consistent human avatar rendering.
• Collaboration with and support of other researchers across various disciplines.
• Communication of research agenda, progress and results.
• If successful, write up your results for possible publication.

• Currently has, or is in the process of obtaining, a PhD in the field of Computer Science, Computer Vision, Machine Learning or related field.
• Must obtain work authorization in country of employment at the time of hire, and maintain ongoing work authorization during employment.
• Understanding of computer vision, 3D geometry, neural rendering, human modeling.
• Experience with deep learning frameworks like PyTorch.

• Knowledge of state-of-the-art human face and hair representations.
• Experience with fine-tuning generative models for controllable generation.
• Intent to return to degree-program after the completion of the internship.
• Proven track record of achieving significant results as demonstrated by publications at leading conferences such as CVPR, ECCV/ICCV, NeurIPS, SIGGRAPH, or similar.