• Όλες οι κατηγορίες
• Ενδοπολυτεχνικές Ανακοινώσεις και Συζητήσεις
• Ακαδημαϊκές Ανακοινώσεις
• Παρουσίαση Διπλωματικής Εργασίας - Σπανού Ειρήνης

Παρουσίαση Διπλωματικής Εργασίας - Σπανού Ειρήνης

• 1

• Συντάχθηκε 08-08-2012 14:12 από Garantonaki Argyro

  Email συντάκτη:

  Ενημερώθηκε: -

  Ιδιότητα: -.
  
  ΠΟΛΥΤΕΧΝΕΙΟ ΚΡΗΤΗΣ
  Τμήμα Ηλεκτρονικών Μηχανικών & Μηχανικών Υπολογιστών
  
  ΠΑΡΟΥΣΙΑΣΗ ΔΙΠΛΩΜΑΤΙΚΗΣ ΕΡΓΑΣΙΑΣ
  
  Ειρήνης Σπανού
  
  με θέμα
  
  “Scene Change Detection-Based Markov Chain Models for Improved Video Traffic Modeling”
  
  [Τετάρτη 22 Αυγούστου 2012, 10:30 πμ]
  [Αίθουσα 145.Π42, Κτίριο Επιστημών, Πολυτεχνειούπολη]
  
  Εξεταστική Επιτροπή
  
  Επ. Καθ. Πολυχρόνης Κουτσάκης (επιβλέπων)
  Καθ. Μιχάλης Πατεράκης
  Καθ. Μίνως Γαροφαλάκης
  
  Abstract
  
  With more than 16 billion videos streamed on YouTube during last May and recent estimates by Cisco that mobile video traffic will increase 25-fold between 2011 and 2016, there is a pressing need to adequately serve large numbers of simultaneous online video transmissions. Network providers need to be able to guarantee the strict Quality-of-Service (QoS) requirements of real-time variable bit rate (VBR) video users, and a good statistical model for multiplexed video traffic can help significantly to evaluate and enhance network performance under various video loads.
  This work proposes and evaluates a new hybrid video traffic model.
  The video traces examined are MPEG-4 sequences and their genre fluctuates between lectures and cartoons, talk shows and action movies, sci-fi and sports. We first build a Discrete Autoregressive model of order one (DAR(1)) and discuss its efficiency on capturing the behavior of actual traces.
  We then proceed to build and evaluate our hybrid model, which combines the DAR(1) model with scene change detection and scene classification techniques. A scene typically consists of a few tens or hundreds of video frames that depict a particular real-world scene and can also be interpreted as a collection of related Groups-of-Pictures (GoPs). The classification is performed based on the average number of bits generated during the scenes and the scene activity is modeled by a Markov chain where each state represents the degree of activity (high/low).
  Finally, we compare the results of the two models and discuss the contribution of our proposed model.
  
  

  ---

  Συνημμένα:

  • DiplomatikiErgasia-Anakoinwsi-Spanou.doc
    Μεταφορτώσεις: 150, Μέγεθος: 43 KB application/msword

• 1