Your responsibilities
- Research: Conduct cutting-edge research on incorporating body language, expressions, active listening, and other conversational features into audio-driven human avatars (in collaboration with one of the leading research labs in the world)
- Development: Develop the most hyper-realistic real-time-conversational audio-driven streaming human avatar model in the world
- Collaboration: Take full ownership of your work and collaborate with MLEs and other engineers to bring your models to production and improve them based on observed production behavior
- (Maybe) Dataset Collection: Work with capturing rigs to gather a one-of-a-kind expressive human dataset for building body language and other conversational features into avatars
- (Maybe) Publication: Publish results in leading venues like CVPR / ICCV / …