Resource-Constrained Object-Based Compression for Surveillance Systems

Add to your list(s) Download to your calendar using vCal

• Adam Powell (Imperial College London)
• Monday 28 February 2011, 11:30-12:00
• Level 9 Mahanakorn Lab., EE Dept..

If you have a question about this talk, please contact Grigorios Mingas.

Current image and video compression techniques, despite being widespread, contain a number of assumptions that make them ill-suited for certain applications. One of these assumptions generates problems in resource-constrained situations. More specifically, the output generated by a compression algorithm does not have to deal with changing memory constraints after it has been encoded. This talk features the design and development of an object-based resource-aware compression system for use in embedded surveillance situations. By abstracting out of the pixel level and into the object level, different techniques and challenges are encountered. The aim of this system is to explore the trade-off of resource usage and quality of the output when the amount of data to be processed is unknown. To adapt to changing memory conditions, the system must have a method to efficiently reduce the amount of memory it takes to store a given object. By freeing memory as limits are reached, it allows for a graceful degradation of quality over time while still maintaining useful output.

This talk is part of the CAS Talks series.

This talk is included in these lists:

• Andrea Picciau's list
• CAS Talks
• Circuits and Systems Group: Internal Seminars
• Level 9 Mahanakorn Lab., EE Dept.

Note that ex-directory lists are not shown.