OpenSpace is looking to improve the core pieces of their mapping and localization capabilities by accurately localizing images taken on construction sites and generating 3D reconstructions.
Requirements
- Experience with camera models and camera calibration
- Comfort with 3D pose representations and transforms
- Experience with at least one of the following: SLAM, Structure from Motion (SfM), 3D reconstruction, photogrammetry, Visual Inertial Odemetry (VIO)
- Experience with 360/spherical image representations
- Familiar with all basic CV concepts like camera models, feature detection and matching, 3D pose representations and transforms, image processing, etc.
- Solid Python programming skills
- 3D visualizations and debug interfaces (OpenGL/PyQt)
- Numerical optimization
Responsibilities
- Design and architect complex, multi-component computer vision systems
- Prototype such systems and run detailed tests to evaluate performance on real world data
- Work with platform engineers to develop production level implementations
- Oversee dataset creation, curation, and labeling as necessary
- Monitor system performance in production and ensure adequate performance
- Assist in debugging and resolving failure edge cases in our production processing pipeline
- Contribute to defining development processes, best practices, and documentation
Other
- M.S./Ph.D. in related field or equivalent industry experience
- Minimum of 2 years industry experience developing Python code
- LI-Remote
- OpenSpace welcomes employees from varied backgrounds and walks of life, and it's reflected in our diverse community
- OpenSpace is proud to be an equal opportunity employer
