Skip to content

eldadoh/poster-detection-using-homography-and-template-matching

Repository files navigation

PROJECT NAME : poster_detection_homography_and_template_matching

LAST UPDATE : 21/06/21

MAIN TASK : 1. Detect (classify and localize) templates (posters) on realograms (real scene)
            using keypoints+homography (poster_detection.py)
            
            2. in case the template doesnt have enough valuable key points, homography will fail to match between the template
               and it place in the scene , so we decided to try template matching algorithm on different scales , 
               and by analyzing correlation map scores for each scale (5 per template) and each template class (currently only 4 'hard' templates)
               (scaled-template-matching.py)

METHODS USED IN THIS PROJECT: key points extractors , homography , template matching , region props , binary image segmentation , 
                               region charcaristics (bbox area , region area , number of points , etc)

ASSETS: - short presentation created for weekly CV meeting 
        - jupyter notebook with couple of experiments 

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published