sportsbitenews/geotagging

Contains the implementation of algorithms that estimate the geographic location of media content based on their content and metadata. It includes the participation in the MediaEval Placing Task 2013.

GeoTag

This is the implementation of the CERTH team's project in the Placing task of MediaEval Benchmarking 2013. The project's paper can be found here.

Main Method

This is an tag-based method, in which a complex geographical-tag model is built from the tags and locations of the images of the training set, in order to estimate the location of each query image. The implemented approach comprises three steps.

A. Filtering: remove noisy and irrelevant tags from the training data, such as machine-tags, and then remove from the training set those images with no tags left.

B. Spatial clustering and local LDA: cluster the training set images based on their location, using k-means on their latitude-longitude values. For each such area that is created, the LDA algorithm is applied to derive a local topic distribution (Local LDAs).

C. Creating bag-of-excluded-words (BoEW): create a set of non-geographic tags, i.e. tags that should not be taken into account for geotagging.

The procedure for the final estimation of the location of the query images is the following.

• Filter the tags of the image that are either machine-tags or belong to the BoEW.
• Compute the Jaccard similarity between this set of tags and the set of tags for each topic of each local LDA.
• Assign the query image to the area with either the highest similarity with any local topic or the highest mean similarity of the topics of each area.
• Determine the most similar training images using Jaccard similarity and use their center-of-gravity.

Instructions

In order to make possible to run the project you have to set all necessary argument in the file config.properties.

Input File Format The dataset's records, that are given as training or test set, have to be in the following format.

			imageID, userID, lat, lon, tags

imageID: the ID of the image.
userID: the ID of the user that uploaded the image.
lat: image's latitude.
lon: image's longitude.
tags: image's tags.

Used Libraries Also, you need the libraries bellow:
1. K-means
2. LDA

Contact for further details about the project

Giorgos Kordopatis-Zilos ([email protected])
Symeon Papadopoulos ([email protected])