This course focuses on visual geo-localization and location-aware image understanding. The learning outcomes include understanding approaches to visual geo-localization, forming input graphs for localization, and segmenting geo-localization. The course teaches skills such as image matching, trajectory reconstruction, and business recognition. The teaching method includes lectures on various topics related to geo-localization and image understanding. The intended audience for this course is individuals interested in computer vision, image processing, and geo-spatial technologies.
Visual Geo-Localization and Location-Aware Image Understanding
University of Central Florida via YouTube
Overview
Syllabus
Intro
Ground Level Imagery
Geo-localization Using Image Matching
Approaches to Visual Geo-localization
Dissertation
Outline
Chapter 3
Geospatial Pruning
Quantitative Localization Results
Ambiguity of local features
Using Multiple Nearest Neighbors
Forming Input Graph to GMCP
K-partite Complete Input Graph
Generalized Minimum Clique
Multiple group of matching images
Robustifying the distance function
Local Search
Geo-localization Results
Robustification Function
Chapter 5
Segment Geo-Localization
MST-based Trajectory Reconstruction
MST Based reconstruction results
Quantitative Results
Error in GPS tags
Triplet Location Estimation
Random Walks on GPS estimations
Experiments
Adaptive Damping Factor Evaluation
Chapter 7
Visual Business Recognition
Multi-Hypotheses Character Recognition
Fusing Text Processing and Image Matching
Summary
Directions for Future Work
Taught by
UCF CRCV
