The Enhancing Computer Vision for Public Safety Challenge is an open innovation competition from NIST PSCR focused on advancing the capacity of no-reference (NR) metrics and computer vision algorithms to support public safety missions. During this 2-phase challenge, NIST PSCR will award up to $240,000.
PSCR seeks solvers to create image or video datasets that depict camera capture problems, such as grime on the lens or sun flare, that cause issues for computer vision applications. These training datasets will help further image quality research and development aimed at better diagnosing and predicting camera problems. This supports PSCR’s mission by expanding the research community and raising awareness of the public safety use case and NIST’s research in this area.
Congratulation to the following teams selected to move forward to Phase 2 of the NIST Enhancing Computer Vision for Public Safety Challenge! The six winners of Phase 1 will each receive a cash prize award and are invited to participate in Phase 2 of the challenge. The Phase 1 Winners are:
Team iAI Tech-NJIT
A roadblock to the deployment of computer vision and video analytics is the myriad problems cameras experience when deployed in real world environments. PSCR wants to stimulate research in the unique area of image quality assessment to support public safety.
To accomplish this, new training datasets and innovative algorithms – referred to as a NR metric for image quality assessment (IQA) or video quality assessment (VQA) – are needed to enhance computer visions systems’ ability to predict and resolve image quality problems in first responder scenarios. Utilizing the NR metrics, deployed systems can implement complex strategies to solve these image quality problems before they even reach public safety analytics.
At the contestant’s discretion, these images and videos will be freely distributed via NTIA’s Consumer Digital Video Library (CDVL.org), providing the research community with valuable datasets for R&D. In the future, successful challenge solutions will be able to identify camera quality problems so that analytics researchers can use these computer resources instead of trying to diagnose and predict these problems manually. The winning solution would also eliminate roadblocks preventing the transition from the development to deployment of new public safety analytics features.
The Computer Vision Challenge is no longer open for submissions. To follow challenge updates and learn more about the challenge, please visit www.pscompvischallenge.com.