Role Overview
We are looking for a Computer Vision Engineer to work on our existing Python-based 3D reconstruction platform. The role involves improving image processing, marker detection, camera calibration, triangulation, and overall reconstruction accuracy.
Responsibilities
- Maintain and enhance the existing computer vision pipeline.
- Improve marker/sticker detection, tracking, and image correspondence.
- Work on camera calibration, camera pose estimation, and multi-view geometry.
- Develop and optimize 3D triangulation and reconstruction algorithms.
- Analyze and reduce reconstruction and reprojection errors.
- Debug and improve performance, accuracy, and reliability of the system.
- Collaborate with backend and frontend teams to integrate algorithm updates.
- Document technical solutions and maintain code quality.
Requirements
- Strong Python programming skills.
- Hands-on experience with OpenCV, NumPy, and SciPy.
- Good understanding of Computer Vision, Camera Calibration, Multi-View Geometry, 3D Reconstruction, Triangulation, Image Processing, and Numerical Optimization.
- Experience debugging and improving existing codebases.
- Familiarity with Git and collaborative development workflows.
Nice to Have
- Photogrammetry or Structure-from-Motion (SfM).
- Bundle Adjustment and optimization techniques.
- 3D visualization tools such as Plotly or Matplotlib.
- Experience with body measurement, garment, or human-body reconstruction systems.
Skills
- Python
- OpenCV
- NumPy
- SciPy
- Git