2015 Fiscal Year Research-status Report
Real-time Video Analytics in Compressed Domain for Big Data Traffic Management
| Project/Area Number |
15K21599
|
|---|
| Research Institution | Waseda University |
|---|
Principal Investigator |
劉 志 早稲田大学, 理工学術院, 次席研究員 (90750240)
|
|---|
| Project Period (FY) |
2015-04-01 – 2018-03-31
|
| Keywords | compress domain / video analysis / traffic management / big data |
|---|
| Outline of Annual Research Achievements |
The growth of the big data is outpacing the cost decline of storage disks and the rate of increase of data network bandwidth. An overwhelming proportion of collected data is visual media such as images &videos. It becomes necessary to filter out the unnecessary video. The inherent problem is how to judge whether the video is important or not. In this fiscal year, we proposed a method to help judge the video important and now we are working on the experiments to verify its performance. Meanwhile, we proposed new schemes for the big data network traffic management. One typical work is that we propose a new smart caching scheme in CCN which could significantly reduce the network traffic.
|
| Current Status of Research Progress |
Current Status of Research Progress
2: Research has progressed on the whole more than it was originally planned.
Reason
As mentioned in the application, the main tasks of the first year are the importance evaluation and the importance calculation based on the compressed-domain information. During the previous fiscal year, I proposed a method to judge the importance and a method to help calculate the video semantics based on the compressed-domain information. Meanwhile, I also proposed methods to help reduce the big data network traffic such as using smart caching in CCN, which is complementary.
|
| Strategy for Future Research Activity |
In this fiscal year, I will continue the experiments on the video semantics calculation based on the compressed domain information, I expect to get the proposed scheme’s results in this year. Meanwhile, I will also keep on proposing new schemes for the big data traffic management besides the compressed domain video analytic.
|
| Causes of Carryover |
Planned purchase are rescheduled in year 2016, hence the money of Y2015 will be switched to Y2016.
|
| Expenditure Plan for Carryover Budget |
The budget will be used in this year according to the plan and the necessity.
|
