Real-time Video Analytics in Compressed Domain for Big Data Traffic Management

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K21599
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Multimedia database; Information network
• Research Institution: Shizuoka University (2017); Waseda University (2015-2016)
• Principal Investigator: Liu Zhi 静岡大学, 工学部, 助教 (90750240)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: ビ ッ グ デ ー タ / マ ル チ メ デ ィ ア 情報処理 / ネ ッ ト ワ ー ク 構成 運用管理評価技術 / ト ラ フ ィ ッ ク 管理 / マ ル チ メ デ ィ ア / 情報ネ ッ ト ワ ー ク

Outline of Final Research Achievements

Vast amount of data (also known as Big Data) is now continuously collected across geographical locations for massive data analysis towards semantically meaningful information and trends. An overwhelming proportion of collected data is visual media such as images & videos. To keep up with the torrid pace of big data growth and avoid paralyzing network congestion and/or data overflow at cloud storage, a smarter data collection scheme, one that targets visual media specifically, must be designed to filter out in real time uninformative data at network edges before it enters the network core.
In this project, we conducted the compressed domain video analysis and designed the video transmission methods in various types of networks. The related results are published or submitted to number of journal and conference papers.

Free Research Field

情報学