Job Title: Computer Vision Researcher – 3D Reconstruction from Video
Role Overview
As a Computer Vision Researcher, your primary focus will be on designing, implementing, and evaluating novel methods for 3D scene reconstruction from video, including dense geometry, dynamic scenes, and real-time pipelines. You’ll collaborate with a multidisciplinary team of researchers and engineers to publish groundbreaking work, integrate your research into production systems, or contribute to open-source frameworks.
Key Responsibilities
- Conduct original research in 3D reconstruction, SLAM, structure-from-motion (SfM), or neural rendering from video data.
- Develop and evaluate algorithms for dense and sparse 3D reconstruction using monocular, stereo, or multi-view video.
- Implement robust, scalable prototypes in Python, C++, or PyTorch/TensorFlow.
- Stay up to date with the latest advances in computer vision and deep learning.
Preferred Qualifications
- PhD or MSc in Computer Vision, Machine Learning, Robotics, or a related field (or equivalent industry experience).
- Strong track record of research in 3D vision: multi-view stereo, SfM, neural fields (e.g., NeRF), or SLAM.
- Experience working with video-based data for geometry estimation or scene understanding.
- Proficiency in Python and C++, with experience in deep learning frameworks such as PyTorch or TensorFlow.