You will work on creating ground truth data tools for all computer vision teams at Magic Leap. To achieve this you will work with motion capture systems, cameras, and robots. You will work within the Perception Products team to deliver novel ground truth tools to enable the next generation of spatial computing.
- Collaborate cross-functionally to develop and optimize calibration for multi-camera systems across a wide range of sensors and environmental conditions
- Analyze ground truth systems and data, especially image data and 6dof transformations
- Develop algorithms and implement SW components for data capture and processing
- Build tools to create ground truth datasets using whatever means possible. Thinking outside the box in terms of software / electrical / mechanical / labeling / off-the-shelf systems to create ground truth systems as rapidly as possible.
- Validate the precision / accuracy of created systems so data is of known quality.
- 5+ years of development experience in C/C++ within an industrial environment
- Experience working in a Linux/unix environment
- Solid understanding of camera models, calibration methods, distortion models and rectification methods, and multi-camera synchronization
- Hands-on experience developing extrinsic calibration algorithms for single and multi camera solutions
- Experience writing maintainable, reusable code, leveraging test driven principles to develop high quality modules
- Experience with non-linear optimization methods and programming libraries
- Experience with calibration between point clouds and camera-based systems
- Experience with Python and/or Matlab
- M.S. in CS, ECE or related fields. Ph.D. preferred.
- All your information will be kept confidential according to Equal Employment Opportunities guidelines.