Real-time video analysis and personalized media streaming environments for large scale athletic events

Authors:
Nikolaos Papaoulakis;Nikolaos Doulamis;Charalampos Patrikakis;John Soldatos;Aristodemos Pnevmatikakis;Emmanuel Protonotarios
Affiliations:
National Technical University of Athens, Athens, Greece;National Technical University of Athens, Athens, Greece;National Technical University of Athens, Athens, Greece;ATHENS INFORMATION TECHNOLOGY, PEANIA, Greece;ATHENS INFORMATION TECHNOLOGY, PEANIA, Greece;National Technical University of Athens, Athens, Greece
Venue:
AREA '08 Proceedings of the 1st ACM workshop on Analysis and retrieval of events/actions and workflows in video streams
Year:
2008

Citing 7
Cited 2

Learning Patterns of Activity Using Real-Time Tracking

IEEE Transactions on Pattern Analysis and Machine Intelligence
Face Recognition: Features Versus Templates

IEEE Transactions on Pattern Analysis and Machine Intelligence
Authentication platform for seamless handover in heterogeneous environments

ICAS-ICNS '05 Proceedings of the Joint International Conference on Autonomic and Autonomous Systems and International Conference on Networking and Services
A breadboard architecture for pervasive context-aware services in smart spaces: middleware components and prototype applications

Personal and Ubiquitous Computing
Bayesian Filtering for Location Estimation

IEEE Pervasive Computing
A generative approach to audio-visual person tracking

CLEAR'06 Proceedings of the 1st international evaluation conference on Classification of events, activities and relationships
On-line retrainable neural networks: improving the performance of neural networks in image analysis problems

IEEE Transactions on Neural Networks

Personalized production of basketball videos from multi-sensored data under limited display resolution

Computer Vision and Image Understanding
A friendly location-aware system to facilitate the work of technical directors when broadcasting sport events

Mobile Information Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents the architecture of the My-e-Director 2012, an FP7 call-1 European funded project in the objective area of Networked Media. My-e-Director 2012 is a unique interactive broadcasting platform, which enables end-users to have access on innovative services, like selection of focal actors/scenes and points of interest within real-time broadcasted streams. Emphasis is placed on the ambient camera selection module, which is the heart of the My-e-Director project. The ambient camera selection is responsible for detecting athletes in large-scale Olympic events. Then, this information is fed back to the system in order to allow for personalized video steaming and delivery services. Since detection of human in athletic event is a very demanding task, our method is based on a retrainable neural network architecture, which non-linearly models the color and texture properties of the object of interest. The retraining algorithm adapts the neural network model to fit the current environmental conditions, while simultaneously trusts as much as possible the previous information. The neural network classifier is combined with a conventional motion-based tracking scheme, which provides accurate contour detection. Aspects of multiple camera configurations are discussed.