← All projects
Computer Vision · Visual Retrieval
Visual Place Recognition
Recognizing places across viewpoint changes and challenging indoor environments.
Research directionOverview
Work on visual place recognition with an emphasis on indoor environments, scalable dataset preparation, and global image descriptors. The project explores adapting retrieval recipes to large collections of indoor scans.
Methods & topics
Highlights
- Preparing an indoor VPR dataset from Matterport-style panoramic captures.
- Investigating descriptor learning and retrieval for indoor environments.
- Designing geographically and spatially meaningful evaluation splits.