INF5090 - Autumn 2015

Lab assignments

Please find below the list og possible projects. We send you an invitation to a doodle poll where you must indicate your preferences which we try (!) to fulfill.

You must indicate your preferences by Thursday 23:59. Those that have not done this will not get a project and thus an F for this part of the course!

Project 1: Obtaining a Video Dataset from YouTube via DASH

Contact Person: Benjamin Guthier (guthier@pi4.informatik.uni-mannheim.de)

Description

Multi-modal video analysis considers the video itself and the surrounding metadata at the same time to extract information from the video [1]. In online video databases, such metadata could be ratings, comments or information about the uploader. In order to test video analysis algorithms, a large corpus of videos and metadata is necessary. This corpus should be a representative sample of all available videos (within chosen parameters), instead of being biased towards a limited set of uploaders or categories [2].

The goal of this topic is to develop a tool that downloads videos and metadata from the online video platform YouTube using the API [3]. It should be possible to set search parameters and specify the videos to be downloaded in an easy way. The videos and metadata should be saved in a format that is easy to load into other tools (e.g., CSV).

YouTube uses Dynamic Adaptive Streaming over HTTP (DASH) for streaming of the videos [4][5]. So investigating the way in which DASH is used by YouTube and enabling DASH-based downloading of videos is another aspect of the topic.

Task

1. Create a tool to download videos and metadata from YouTube
2. Develop a strategy to select a number of independent videos to download
3. Analyze how DASH is used by YouTube to provide multiple versions of a video and to allow switching during playback
4. Evaluate the performance of the tool
5. Write a 10 page report (IEEE style) which discusses the DASH protocol and the implemented tool

Skills

JAVA or C++ programming skills

References

[1] Atrey, Pradeep K., et al. "Multimodal fusion for multimedia analysis: a survey." Multimedia systems 16.6 (2010): 345-379.

[2] Cao, Juan, et al. "MCG-WEBV: A benchmark dataset for web video analysis." Beijing: Institute of Computing Technology 10 (2009): 324-334.

[3] http://developers.google.com/youtube/v3/

[4] Stockhammer, Thomas. "Dynamic adaptive streaming over HTTP - standards and design principles." Proceedings of the second annual ACM conference on Multimedia systems. ACM, 2011.

[5] ISO/IEC 23009-1:2012. Information technology - Dynamic adaptive streaming over HTTP (DASH) - Part 1: Media presentation description and segment formats.Project 2: An NFC-enabled Learning Game for the Topic of Computer Networks

Contact Person: Philip Mildner (mildner@informatik.uni-mannheim.de)

Description

Near Field Communication (NFC) enables devices to communicate in a small scope of only a few centimeters. This physically oriented scheme allows for personal and spontaneous communication scenarios that do not require a specific infrastructure to run.

The goal of this project is to evaluate the capabilities of NFC in the course of a learning game about networking topics. Therefore, mobile devices should communicate via NFC tags. The devices depict nodes in a virtual network and NFC tags act as messages that are sent between the stations. The goal of the game is to simulate certain networking protocols/algorithms such as RIP, TCP, Dijkstra or CacheCast. Players have the task to manually reenact the protocol by passing NFC tags through the virtual network. Game elements could be to have certain scenarios the players have to solve or time trials where players have to act as fast as possible.

The game should be implemented on the Android platform. The finished game should then be evaluated with regard to the performance of the NFC capabilities.

Task

1. Create a game concept and choose appropriate networking algorithms to use in the application.
2. Implement the game on the Android platform.
3. Evaluate the performance of the game and the NFC capabilities.
4. Write a 10 page report (IEEE style) which discusses the implemented game.

Skills

JAVA/Android programming skills

References

[1] Vitor Subtil. 2014. Near Field Communication with Android Cookbook. Packt Publishing.
[2] Gregor Broll, Roman Graebsch, Paul Holleis, and Matthias Wagner. 2010. Touch to play: mobile gaming with dynamic, NFC-based physical user interfaces. In Proceedings of the 12th international conference on Human computer interaction with mobile devices and services (MobileHCI '10). ACM, New York, NY, USA, 459-462. DOI=10.1145/1851600.1851706
[3] Avinash Nandwani, Paul Coulton, and Reuben Edwards. 2011. NFC Mobile Parlor Games Enabling Direct Player to Player Interaction. In Proceedings of the 2011 Third International Workshop on Near Field Communication (NFC '11). IEEE Computer Society, Washington, DC, USA, 21-25. DOI=10.1109/NFC.2011.19
[4] Luis F. G. Sarmenta. 2012. Tangible and casual NFC-Enabled mobile games. In Proceedings of the 10th international conference on Pervasive Computing (Pervasive'12), Judy Kay, Paul Lukowicz, Hideyuki Tokuda, Patrick Olivier, and Antonio Krüger (Eds.). Springer-Verlag, Berlin, Heidelberg, 361-369. DOI=10.1007/978-3-642-31205-2_22
[5] Squire, K. and Jenkins, H. 2003. Harnessing the power of games in education. Insight. 3, (2003), 5–33.

Project 3: Evaluation of Streaming Performance in Wireless Mesh Networks