Computer Vision Geometry Engineer

Location Santa Clara
Contact email: ari@deepabacus.com
Job ref: 2
Published: about 1 month ago

Company:

  • Technology company who are evolving and democratizing ADAS 

  • Led by engineers from Tesla’s core Computer Vision team 

  • Opportunity to get in as they go through their Series C funding round

What you’ll do: 

  • Integrate and implement real-time mapping and localization algorithms for outdoor applications.

  • Estimate the 3D location and trajectory of traffic objects in a monocular camera setup.

  • Simulate and test the algorithm in a ROS (Robot Operating System) environment.

  • Deploy a computationally optimized system and achieve real-time performance at embedded target platforms.

  • Explore a new approach for visual SLAM systems leveraged from deep learning.

What you’ll need to succeed: 

  • MS or Ph.D. degree in Computer Science, Electrical Engineering, or equivalent field.

  • 5+ years of working experience in relevant fields such as computer vision, machine learning, and robotics.

  • Strong programming skills in C++ and experience with Python.

  • Solid understanding of an entire pipeline of real-time visual SLAM, including camera model, feature tracking, epipolar geometry, triangulation, and bundle adjustment.

  • Wide experience with OpenCV and open-source SLAM libraries.

  • Experience with bundle adjustment optimization libraries (e.g. g2o or ceres).

  • Experience with sensor fusion with multiple sensors such as camera, odometry, IMU, radar, lidar, etc.

  • Experience with ROS or other alternative middlewares.

  • Experience with utilizing deep learning-based object detection and segmentation.