We are seeking a highly skilled Research Scientist with a strong background in computer vision, computer graphics, and machine learning. This role is ideal for candidates passionate about generating high-quality images and advancing the state of the art in human performance and scene relighting, and material decomposition. You will be at the forefront of developing new algorithms that drive the future of visual storytelling.
Key Responsibilities- Conduct applied research in computer graphics and machine learning - focused on high-fidelity reconstruction and photorealistic image synthesis to solve real challenges in VFX and content creation pipelines.
- Architect, execute, and assess innovative algorithms and generative models or pipelines for image and video synthesis, relighting, illumination estimation, physically based material decomposition, and (Dynamic) Gaussian Splatting, ensuring a viable transition from initial prototype to production deployment.
- Lead high-quality data acquisition and curation efforts to support next-generation machine learning algorithms for VFX and virtual production.
- Engage directly with VFX artists, supervisors, and production teams to understand creative and technical pain points, ensuring research efforts are grounded in real production needs.
- Contribute to and stay at the forefront of the research community by sharing findings, collaborating with peers, and presenting work at top-tier conferences (SIGGRAPH, CVPR, NeurIPS, ICCV, ECCV, ICLR).
- Collaborate closely with engineering, production, and business teams to translate research outcomes into production-ready solutions, bridging the gap between research prototypes and real-world VFX workflows.
- Identify and champion new research directions that can have significant impact on content creation and virtual production.
Required Qualifications- MSc with relevant research experience, or PhD, in Computer Graphics, 3D Computer Vision, Machine Learning, or a closely related field.
- Demonstrable research experience in 3D Computer Vision and Computer Graphics, with contributions to major venues (e.g., SIGGRAPH, CVPR, NeurIPS, ICCV, ECCV, ICLR) through publications, technical talks, or open-source work.
- Deep expertise in computer graphics and machine learning, with hands-on experience in areas such as image and video synthesis, relighting, illumination estimation, physically based material decomposition, photorealistic rendering and (Dynamic) Gaussian Splatting.
- Familiarity with modern generative architectures - diffusion models (image/video), transformer-based methods, or demonstrably transferable experience.
- Comfortable working at both the algorithmic and implementation levels - from model design and optimization to large-scale data processing and integration into production systems.
- Track record of translating research into production-quality solutions, navigating real-world constraints such as data limitations, performance requirements, and production timelines.
- Experience working directly with artists, engineers, or production teams to ensure research outputs meet the quality and usability bar for creative workflows.
- Strong attention to detail in image and signal processing, with a drive to achieve high-fidelity, visually compelling results.
- Exceptional communication and collaboration skills, with the ability to work effectively across research, engineering, and creative teams in a fast-paced environment.
- High level of ownership, initiative, and a positive, problem-solving attitude.
Preferred Qualifications- Passion for movies, TV, and storytelling.
