Intern: Software Engineering, Multi-view 3d Reconstruction

Intrinsic

Munich, Germany
On-site
C++ and/or python
3d computer vision concepts
Structure from motion (sfm)
Develop and implement a dense multi-view reconstruction pipeline using posed images to generate accurate 3D ground truth for single shot reconstruction algorithms

Job Summary

  • Develop and implement a dense multi-view reconstruction pipeline using posed images to generate accurate 3D ground truth for single shot reconstruction algorithms.
  • Apply Deep learning based Structure from Motion (SfM) techniques to create a robust benchmark for evaluating point-cloud models.
  • Collaborate with the perception team to validate ground-truth pipelines and support the direct evaluation of advanced perception algorithms.

Matching Summary

Develop and implement a dense multi-view reconstruction pipeline using posed images to generate accurate 3D ground truth for single shot reconstruction algorithms.

Skills & Requirements

Must-have

  • C++ and/or Python
  • 3D computer vision concepts
  • Structure from Motion (SfM)
  • Multi-View Stereo (MVS)
  • point cloud processing
  • 3D vision libraries

Nice-to-have

  • AI robotics
  • industrial robotics
  • deep learning techniques
  • benchmarking and evaluation

Key Requirements

  • Bachelor's student in CS, Robotics, CV
  • Master's students preferred
  • Experience with SfM, MVS, or photogrammetry
  • Familiarity with 3D geometry and linear algebra
  • Experience with COLMAP, Open3D, PCL

Work Rights

Not specified

Tailored Resume

Cover Letter