Research Scientist - Generative AI Computational Photography, Reality Labs

Research Scientist - Generative Ai Computational Photography, Reality Labs

Meta Reality Labs is at the forefront of innovation, bringing together a dynamic team of researchers, developers, and engineers dedicated to advancing AR and VR technologies. Our Core AI group is seeking a Research Scientist to tackle cutting-edge research challenges in image restoration using Generative AI methods, such as diffusion and autoregressive approaches. This role aims to push the boundaries of state-of-the-art research in image restoration, enhancing the quality and realism of visual content.

Responsibilities

Conduct pioneering research and develop novel Generative AI models, focusing on image restoration tasks such as image and video super-resolution, denoising, deblurring, or frame interpolation.

Publish research findings in top-tier journals and present at leading international conferences.

Collaborate with cross-functional teams throughout the project lifecycle, from initial research to prototyping and deployment.

Work closely with product partners to integrate research outcomes into Meta's products, enhancing user experiences.

Minimum Qualifications

Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience.

PhD in Computer Vision, Machine Learning, or a related field, with a focus on Generative AI and image restoration.

Advanced background in designing machine learning algorithms, including experience with training and inference of novel ML architectures.

Proven experience in cross-functional collaboration and teamwork.

Preferred Qualifications

Experience as a software engineer through internships, work experience, coding competitions, or contributions to widely used open-source repositories (e.g., GitHub).

Demonstrated success in achieving significant research outcomes, evidenced by grants, fellowships, patents, and first-authored publications at leading conferences such as CVPR, ECCV/ICCV, NeurIPS, ICLR, or SIGGRAPH.

Solid mathematical foundation and understanding of numerical optimization, linear algebra, and probabilistic estimation.