Collaborative Content Generation Architectures for the Mobile Augmented Reality Environment

Presented at the 9th IEEE International Symposium on Mixed and Augmented Reality (ISMAR 2010), Seoul, Korea.

Full paper: http://link.springer.com/chapter/10.1007%2F978-1-4419-9845-3_6

Transcript

2. •  Universidad Politécnica de Madrid: –  Telecommunica6ons Engineer. –  Master of Network Engineering and Telema6c Services. –  PhD student. •  Research Interests: –  Collabora6ve architectures for mobile environments. –  Network Science. –  Recommender systems and clustering algorithms based on social data. –  User Experience. 2 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010
3. Mo#va#on and Context of Use Collabora#ve Content Genera#on Pyramid Using the Taxonomy Conclusions and Future Work 3 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010
5. •  A new age in the mobile world: – Powerful mobile devices. – Design focused on the user, not on the technology. – Integra6on with social networks to allow mobile collabora6on. •  The rebirth of AR: – Change the user experience in mobile applica6ons. – Change the way a developer works: new inputs. 5 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010
6. •  Mobile Collabora6ve AR: – The joint of both areas is a direct outcome. – Our research is focused on systems that track without using wearable technologies or fiducial markers. – The majority of current AR applica6ons are supported by smartphones. •  World wide acceptance •  More comfortable and inexpensive for end users. 6 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010
7. •  Defini6ons and taxonomies are essen6al tools that lead us to a beYer understanding of a research area. •  A concrete characteris6c differs from one applica6on to another in the Mobile Collabora6ve AR environment: –  The way the content to be shown as informa6on overlays is generated. –  The collabora6on capabili6es each applica6on allows, is directly related to this characteris6c. •  Our work is oriented to developers and researchers who need to clarify the context in which they are working. 7 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010
9. •  Collabora6ve content genera6on is divided into 3 levels. •  Each level iden6fies: –  Technologies. –  Architectures. –  User perceived impact. •  Pyramidal structure: –  Each level is supported by the architecture and technology of the levels below. 13/10/2010 Daniel Gallego Vico - dgallego@dit.upm.es 9
10. 10 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010 Applica#on Server Context devices GPS POI Search Area Informa#on overlays •  Centralized content genera6on and management. •  No contribu6on is made by the end user to POI’s informa6on. •  Technological and architectural base for the upper levels. •  Applica6ons based on context awareness.
11. 11 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010 Applica#on Server Context devices GPS POI Search Area Add/request AR overlays Social and Content networks •  The mobile collabora6ve AR starts here. •  Informa6on layers are generated by integra6ng different content sources. •  The end user is able to select a POI and aYach informa6on to it. •  Social awareness.
12. 12 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010 Applica#on Server Context devices GPS POI Search Area Add/request AR overlays Social and Content networks AR genera#on in Real Time •  Informa6on overlays related to a POI generated in real 6me by the users. •  This informa6on could be stored, shared or just can be persistent session. •  Distributed architecture is needed (e.g. P2P) •  A research area to explore.
13. 13 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010 •  The Resources Needed are a combina6on of technological and architectural requirements. •  The User Perceived Impact grows with the collabora6on characteris6c.
15. 15 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010 •  Nearest Tube: – AR map to inform people about where the nearest tube sta6on is. •  Theodolite: – Informa6on about posi6on, al6tude, bearing... •  Based on context awareness techniques. •  Content genera6on in one direc6on: – From the applica6on’s servers to the smartphone user client.
16. •  Low Social level: –  Let users add simple pieces of informa6on to exis6ng POIs. –  The crea6on of POIs is only achievable for developers. –  An average user is not able to generate complete content. •  High Social level: –  Any end user is able to create POIs and to aYach informa6on that will be visible for the rest of the community. 16 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010
17. 17 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010 •  Entertainment: –  Small games or MMOG. –  Enable the interconnec6on of several users to generate AR content in a real 6me and collabora6ve way. •  Educa6on: –  A live lec6on could be given following a 1 to N user connec6on model. –  The group can create AR contents that everybody could see. •  Medicine: –  Live medical distance examina6ons. –  How to treat the illness by genera6ng graphical or textual informa6on over the image sent by the sick user.
19. •  The Collabora#ve Content Genera#on Pyramid… –  … conceptualizes and clarifies the Collabora6ve Mobile AR environment. –  … has 3 levels according to the way the AR content is generated: •  Isolated •  Social •  Live –  … helps developers to understand what type of applica6on they want to achieve, and what the necessary components are. –  … discovers new research areas (e.g. Live level). 19 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010
20. •  Evalua#on of our taxonomy: – Is it useful for the researchers and developers? •  Genera6on of AR content based on cloud compu#ng techniques. •  Crea6on of standard APIs to share POIs in any applica6on and avoid redundancy among them. •  Addi6on of another dimension in the taxonomy to analyze the social impact of using every kind of applica6on. 20 Daniel Gallego Vico - dgallego@dit.upm.es 13/10/2010