#18: MLDublin meets First Data @ DogPatch Labs- 1 min
For meetup #18 we were happy to be sponsored by First Data, thanks to Samanta and the team for the help putting this event together.
Christopher will present what First Data do and share an insight in to what is done with Machine Learning within the company.
Many applications such as autonomous navigation, urban planning and asset monitoring, rely on the availability of accurate information about objects and their geolocations. In our talk we propose to automatically detect and compute the GPS coordinates of recurring stationary objects of interest using street view imagery (Google Street View, Bing Streetside, Mapillary, etc.). The designed computer vision pipeline relies on fully convolutional neural networks to detect objects and a triangulation-based Markov Random Field model to geotag objects. The experiments report GPS accuracies within 2 meters, which is comparable with the precision of single-frequency GPS receivers.
Andrew is a PhD student with the Edinburgh Centre for Robotics working in the area of computer vision. Talk description to be annoucned.