University of Twente Student Theses
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.
Sparse-View camera pose estimation of infrastructure
Hassing, Tijn (2025) Sparse-View camera pose estimation of infrastructure.
|
PDF
58MB |
| Abstract: | Camera pose estimation is a foundational technique for object localization, tracking, and 3D reconstruction, with widespread applications in robotics, augmented reality, and computer vision. This research systematically evaluates both traditional methods, like Structure-from-Motion (SfM), and deep learning–based camera pose estimation methods in sparse-view conditions (2–4 images). We evaluate on the public IMC Phototourism dataset and perform a case study on a real-world bridge under varying lighting and motion-induced blur. By testing on both these elements we found that learned models—specifically VggSfM and RayDiffusion—match or exceed classical SfM not just in rotation and translation recovery, but also in robustness to different environments. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 54 computer science |
| Programme: | Computer Science BSc (56964) |
| Link to this item: | https://purl.utwente.nl/essays/107680 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page