Neural-Assisted Video Coding Intern

Job Description

Join the leader in entertainment innovation and help us design the future. The Dolby U internship program offers impactful, project-based work experience in a collaborative, creative environment where you work side by side with industry leaders. Amplify your insatiable curiosity by implementing real-world solutions that revolutionize how people communicate and how entertainment is created, delivered, and enjoyed worldwide. We offer a collegial culture, challenging projects, and excellent compensation and benefits. For this high functioning team, the nature of the work benefits from frequent in-office presence, and employees may be expected to work from a company office up to four days per week. While we remain committed to flexibility, this team’s rhythm includes regular in-person connection to support collaboration, learning, and impact. For any student seeking to gain invaluable expertise through meaningful, personal contributions, we invite you to join us in continuing to design a future where technology meets entertainment! The Legal Department provides core legal support to all Dolby functions, runs several IP-centric businesses, and leads government relations. We are looking to recruit interns in the video coding standards innovation team. The qualified candidate will develop novel video compression technology using learning-based algorithms to conventional video compression that enable the next generation of Dolby technologies. Responsibilities: * Design and implement innovative neural network/learning-based algorithm for learned transform in conventional video coding. * Document and present the new algorithms and implementations in various forms, such as technical white papers, etc. Qualifications: * Experience with Video Compression standard such as H.265/HEVC, H.266/VVC, and beyond. * In-depth expertise in deep learning (DL) and statistical machine learning (ML) * Extensive programming experience in python/C/C++ applied to image/video coding/processing …