Exploring Relationships Between Ground and Aerial Views by Synthesis and Matching

Wednesday, June 30, 2021 3 p.m. to 5 p.m.

Final Examination of Krishna Regmi for the degree of Doctor of Philosophy in Computer Science

Download Bifold

We have developed deep neural network based architectures to generate cross-view images, e.g. aerial image corresponding to the ground-level scene, and also developed a method to subsequently utilize the features of synthesized images to facilitate geolocalization task employing image-based matching. Such localization is useful in vision based navigation in a GPS denied environment, e,g. a moving car can selflocalize itself in case of GPS failure. Similarly, given the aerial images of a particular region in the world over time, we can analyze the changes (development or destruction) the region undergoes and study its historical evolution. Additionally, the authenticity of the geo-tagged images/videos uploaded in the social media can be validated, which can help purge the fake information from being circulated in the media.

Read More

Location:


Contact:


Calendar:

CRCV

Category:

Academic

Tags:

UCFCRCV