SBIR-STTR Award

Use of No-Reference Measurements of Subjective Quality to enhance Next Generation systems for Video Distribution
Award last edited on: 1/30/2009

Sponsored Program
SBIR
Awarding Agency
NSF
Total Award Amount
$600,000
Award Phase
2
Solicitation Topic Code
-----

Principal Investigator
Pravin Mane

Company Information

VQlink Inc

75 5th Street Nw Suite 222
Atlanta, GA 30308
   (404) 385-3169
   info@vqlink.com
   www.vqlink.com
Location: Single
Congr. District: 05
County: Fulton

Phase I

Contract Number: ----------
Start Date: ----    Completed: ----
Phase I year
2007
Phase I Amount
$100,000
This Small Business Innovation Research Phase I project concerns the design and development of systems for real-time enhancement of the quality of video distributed over IP-based networks, including best-effort links. The proposed research provides a systematic way of realizing the above goal, building upon recent research on the measurement and optimization of video and network subsystems. A central aspect of the proposal is a zero reference method for predicting video quality anywhere in the network, with high correlation with subjective measurements of quality. The research will include the topologically simple example where video quality at a specific place in the network (such as the receiver at the end of the video communication chain) is enhanced by making use of algorithms that are at the core of the system. If successful, the solutions proposed in this research will help content providers, aggregators, distributors and receiver manufacturers in meeting user expectations in emerging systems for entertainment video. This is by itself a significant market. The proposed research will also contribute directly to the science of high-quality visual communications

Phase II

Contract Number: ----------
Start Date: ----    Completed: ----
Phase II year
2008
Phase II Amount
$500,000
This Small Business Innovation Research (SBIR) Phase II project concerns the design, development and commercialization of systems for real-time measurement and enhancement of the quality of video distributed over emerging networks. The Phase I project has demonstrated the feasibility of a real-time version of an automatic video quality meter (AVQ) that has several novel features. By not requiring a reference to the source video, and by being compute-scalable, the AVQ meter can reside anywhere in the video distribution chain. In addition, AVQ scores correlate extremely well with subjective assessments of users, making it valuable not only as an accurate measurement tool, but also as the starting point for VQ Enhancement (VQE). This Phase II work will create an industry-ready suite of AVQ products that can reside in the network (AVQ-N), home (AVQ-H) and Set Top Box (AVQ-STB). It will also create solutions for video quality enhancement (VQE) that will reside in several network points. The AVQ and VQE products will depend on several innovations in design, including the retention of performance in the leaner versions of AVQ and VQE, and portability of a core AVQ engine into multiple PC-native, board-level and chip-level platforms. In rapidly emerging scenarios, video information is generated, gathered and distributed in real time, using a larger number of communication networks than ever before. Increasingly, users who are exposed to the notion of high-quality, high definition television will demand a high quality of video experience in all circumstances, in both fixed and mobile systems. The solutions proposed in this research will initially help content providers, aggregators, distributors and receiver manufacturers in meeting user expectations in emerging systems for entertainment video. These systems include cable and telco services as well as mobile and internet video